Generate Variable Rate Prescriptions Choosing a Computer to Run SlantView

Transcription

1 Table of Contents SlantView User Guide Definitions Install SlantView Compatibility Licensing Basic SlantView Workflow Download Data Process Data View Data Save Data Share Data Automated Statistics Reports Data Upload Advanced SlantView Functionality Adjust Shadow and Soil Filtering Settings Use Custom Filters and Smart Detection Assess Data Quality Adjust Data Processing Settings Exporting Images Stress Analysis Weed Detection Population File Formatting Batch Processing Update Firmware Manually Adjust Boresight Alignment Native Resolution Mosaic Stitching Combine Multiple Flights Calibrated vs. Uncalibrated Imagery Measure Pixel NDVI and Reflectance Use SlantView to Scout a Field

2 Generate Variable Rate Prescriptions Choosing a Computer to Run SlantView

3 SlantView User Guide Your SLANTRANGE system is dependent on intelligent mission planning and operation. We ask that you carefully read this guide before taking to the skies to ensure you get the most out of your system and your harvest. Thank you, The SLANTRANGE Team 3

4 4

5 Definitions The following is an introduction to SLANTRANGE's processing software, SlantView. A few notes: words shown in bold are references to on-screen options in SlantView and file explorer windows, and words shown in italics are common SlantView terms to become familiar with. A list of common and potentially confusing SlantView Terms, in alphabetical order: Color Scale The Color Scale of the Stress Map should be adjusted based on the growth stage of the crops at the time of the flight. A field with a lot of bare soil and little crop content (early stage crops) may initially produce a map that displays a gradient of light red (relatively healthy areas) to dark red (relatively stressed areas). Alternatively, a field with a lot of crop content and little bare soil (mature, closedcanopy crops) may initially produce a map that displays a gradient of light green (relatively stressed areas) to dark green (relatively healthy areas). The measured Stress values, 0.0 to 0.5 do not change, but we recommend adjusting the Color Scale applied to the range 0.0 to 0.5 to produce an intuitive gradient of green (healthy areas) to red (stressed areas) within the Stress Map. Custom Boundary SlantView's map cropping tools can be used to trim map products to a Custom Boundary. Custom Boundaries are most commonly used to remove images taken beyond the edges of a field. Custom Filter Custom Filters (also referred to as Smart Detection) allow users to search fields for unique spectral signatures that they want to locate, measure the Stress of, and/or filter out. Custom Filters have been used to identify noxious weeds, pests, and water leaks among other potential concerns in a field. Custom Filter Region The red box drawn by the user that defines the spectral content the user wants to identify throughout the field. Dataset The folder containing the collection of images and files the SLANTRANGE sensor stores from a given flight. Each time the SLANTRANGE sensor begins taking images as indicated by the AIS light blinking green, a new Dataset will be created in the sensor's memory. For SlantView to correctly Process a 5

6 Dataset, the folder (as viewed in the Windows File Explorer) must be named with a 15 character time and date identifier in the form (year)(month)(day)t(hour)(minute)(seconds) e.g. a Dataset with the title " T was collected on May 30, 2016 at 22:34:23 UTC (Coordinated Universal Time). Filter View Press the F7 key while active in the Image Window to display the Filter View, which shows all the content being filtered out by the Soil, Shadow, and any Custom Filters. Image Window The secondary SlantView window, displays the high resolution images outlined by the white border following the cursor in the Map Window. Map Window The main SlantView window which displays the map products that combine all the images and data from a given Dataset. The majority of the tools for viewing, editing, and exporting the data are found in the Map Window's toolbars and on-screen buttons. NDVI maps Acronym for Normalized Differential Vegetation Index: green, red, and red-edge NDVIs are standardized ratios of spectral reflectance developed to identify live green vegetation, commonly used in the context of remote sensing. _Plant Size_map The Plant Size map is available for data with Population processing applied. It is a relative measure of the area of each plant when viewed from above, meant to show where plants in a field are bigger and growing more quickly, versus smaller and getting a slower start. The colorbar on the left is set such that plants in the 75th percentile are assigned a size value of 1 and a color of green. From there a size of 0.8 means those plants are 80% as large as the plants in the 75th percentile, and 1.2 would be 120% as large as the plants in the green section. Plant size maps can be generated for fields that are in the population measurement stage (bare soil visible between plants), as well as for fields slightly beyond the population measurement stage where some areas of plants have begun to overlap. Population map The Population map is a map product available only with Population Processing. View this map for all information related to plant counting. Plants and weeds are marked in green and orange, respectively. Population Processing Use this Processing option for plant counting. Population Processing should only be applied to fields of young crops where bare soil is exposed inbetween individual plants, and leaves from individual plants are not yet overlapping one another (e.g. V1 to approximately V3 corn). To produce useful map 6

7 products with Population Processing, unique mission planning constraints on flight speed and altitude, aircraft orientation, and lighting and wind conditions must be followed (specifically lower flight altitude and speed) as described in the Mission Planning section of this guide. Process The terms Process and Reprocess as referenced in this guide mean to load a Dataset from scratch. A typical Dataset will Process in 5 to 10 minutes. Once a Dataset is Processed a Workspace is opened in SlantView. Most changes to map settings, including annotations, trimming, Soil and Shadow Filters, Custom Filters, etc. will be applied to the map products upon Recalculating maps, which takes a fraction of the time of Reprocessing maps. The exceptions are changes to Auto flat-field calibration, Auto boresight alignment, aircraft Attitude Filters (only if the Attitude Filters are widened), and the Use compass data and Enhance geolocation settings, which require Reprocessing to take effect. Recalculate The difference between Recalculating maps and Reprocessing Datasets is perhaps the most confusing SlantView terminology. Recalculating maps takes a fraction of the time of Reprocessing Datasets (~1 minute in most Workspaces). Click the Recalculate button in the Map Window toolbar or settings menu to apply changes in Workspace settings to the map products. For example: after making changes to Boresight Adjustments, Soil and Shadow Filter settings, or Custom Filters. Shadow Filter An adjustable filter that removes pixels of dark and shaded content to isolate uniformly illuminated plant material [the only material that is suitable] for accurate and repeatable health measurements and creation of map products. Soil Filter An adjustable filter that removes pixels of bare soil and other non-plant content to isolate plant material for health measurements and map product generation. Stress Analysis Processing The third option for Processing a Dataset, Stress Analysis Processing, is the most common. Use it during all crop growth stages that do not fit the conditions for Weed Detection or Population data products. Apply to all crops that have matured past the Population Processing stage. Stress Map The primary data product displayed in the Map Window measuring the health of crops in the field and identifying areas that may require your attention. In SlantView, Stress is assigned a value between 0.0 (least stressed) and 0.5 (most stressed) on a color scale from dark blue (least stressed) to dark red (most stressed). The Color Scale bar on the left edge of the Map Window shows the Stress values and the color corresponding to each. 7

8 Vegetation fraction map A data product displayed in the Map Window measuring crop density. Weed Detection Processing The first option for Processing a Dataset, Weed Detection Processing should only be applied to unplanted fields or fields with pre-emergent crops. All plant material in the field will be treated as a weed. Weed Detection Map The Weed Detection map is a map product available only with Population Processing. It detects plant content between plants and rows with a different spectral signature than the crops, highlighting the weeds in red on top of the satellite imagery. Workspace Once a Dataset is Processed, the combination of maps and settings active in the SlantView windows is collectively referred to as the Workspace. A saved Workspace includes any annotations, trimming, map products, Soil Filters, Shadow Filters, Custom filters, etc. applied to a Dataset. Note that a Workspace will not save changes to the Color Scale; the Color Scale is native to a particular Map Window. For example: if you Process a Population Dataset and adjust the Color Scale to a green-red gradient for a field with high bare soil content and stress values near 0.4, then Process a Stress Dataset in the same window, you will likely get a map with a light green to dark green gradient because the Stress values of the mature crop will be closer to to the other end of the spectrum at 0.1. Yield Potential map A data product displayed in the Map Window that estimates crop yield potential based on plant health and size. 8

9 Installing SlantView Click here for a quick video tutorial on getting started with SlantView! SlantView, available for download at: is your software tool for processing and viewing data gathered from your SLANTRANGE system. After a flight, connect your sensor to a computer via Ethernet cable (1p, 2i, and 2p) and use SlantView to download and view the data in minutes. If using a 3p, just pull the memord card, copy the files to your hard drive, and begin processing. No highspeed wireless data connection or cloud processing required! To install SlantView, register at analytics.slantrange.com as shown in the top of the Figure. After entering your information, click the Download SlantView box. After downloading, open the SlantView_Install file in the downloads folder, and navigate through the install prompts. After successful installation, the SlantView Map Window will open next to an account information window, as seen in the Figure. Enter the same username and password you used to login at analytics.slantrange.com. 9

10 10

11 Compatibility SlantView is designed exclusively for 64-bit Microsoft Windows operating systems, Windows 7 and later. To prevent any noticeable decrease in your computer's response time while running SlantView, we recommend using a computer with RAM that is approximately double the size of the Dataset(s) open at a given time, e.g. 8GB RAM for a 4GB dataset. Upon successful startup, the SlantView Map Window appears as shown below. Check to see if your computer's speed is limited: on many Windows computers, Balanced or Power Saver modes are default settings when running on battery power. Enabling your machine's version of High Performance mode will allow for faster downloading and Processing. To check this setting, go to Control Panel Hardware and Sound Power Options as shown below. 11

12 Windows high performance mode 12

13 Licensing SlantView Lite SlantView Lite is a free version of SlantView that allows the exporting of processed images for use with 3rd party software. SlantView Basic SlantView Basic includes all the functionality of SlantView Lite, along with SLANTRANGE's unique ability to produce full field Stress and NDVI maps on a standard consumer laptop, with no internet connection, in minutes, not hours. Within SlantView, vegetation content is isolated from shadows and soil, and maps can be trimmed, annotated, and exported to share with collaborators. SlantView Pro SlantView Pro includes all the features of SlantView Basic, along with the ability to view Vegetation Fraction and Yield Potential maps, perform plant population and weed counts, and use Smart Detection to create Custom Filters. SlantView License Features Feature Lite Basic Pro No Network Connection Required X X X Export Calibrated Reflectance Images X X X Instant Coverage/Quality Check X X X Process Images with 20% Overlap X X Rapid In-Field Processing X X Annotations X X Statistics Reports X X High-Resolution Imagery X X NDVI (GNDVI, RNDVI, RENDVI) X X Vegetation Stress X X Population Density X Population Size Distribution X Weed Coverage X Vegetation Fraction X 13

14 Yield Potential User-Defined Smart Detection X X 14

15 Basic SlantView Workflow For 1p, 2i, & 2p systems A typical workflow for downloading, viewing, and sharing data is as follows: 1. Connect the sensor to SlantView, and download the data to your hard drive. 2. Open the Dataset in SlantView and Process as Stress or Population Analysis with default settings. A typical Dataset (between 500 and 1500 images) will take between 5 and 10 minutes to Process with a standard consumer laptop. 3. If the Stress map is predominantly black, dark red, or blue, adjust the Color Scaling of the Map Window by holding CTRL and scrolling the mouse wheel. If using a touch-screen computer, click the up and down arrows next to the Color Scale bar on the left side of the Map Window. Adjusting the field to a balance of green, yellow, and red provides an intuitive display of the gradient from healthy to stressed areas. Refer to the paragraph Example Stress Analysis for MAture Crops on the View Data page for more details on adjusting the Stress Map Color Scaling. 4. Crop the map to the region of interest. 5. Use the various Map Window views to find and assess any stressed areas of the field. 6. Annotate features of interest. 7. Save a KMZ, shapefile, or multi-page GeoTIFF of the maps, and save individual image KMZs and GeoTIFFs if desired. 8. If you will be viewing the data in the future, save your Workspace to save time when reloading the data in SlantView! For 3p & 3PX systems The SLANTRANGE 3p system incorporates a removable SD card, further reducing the time from flight to map generation by eliminating download time in SlantView. Just pull the card from the sensor, insert into your computer, and process in SlantView! Thus, the workflow for the 3p is an abridged version of the workflow for the 1p, 2i, and 2p: 15

16 1. Pull the microsd card from the 3p field sensor and insert into your computer via adaptor or micro SD port, then move the files to your desired data folder. Note that you will have to manually delete datasets from the card within the Windows file explorer to make space for future flights, unless you use SlantView's Organize data tool with the option to Delete from sensor enabled. 2. Follow steps 2 through 8 of the 1p, 2i, and 2p procedure. 16

17 Download Data If using a 3p system Pull the microsd card and insert into your comptuer (depending on your type of computer/tablet you may need an adapter such as microsd-sd, or microsd-usb). You can transfer the data through SlantView using the Organize data option shown below, or copy the files to your desired directory through the traditional Windows file explorer. If you transfer the dataset folders manually, be sure to delete them from the card to make room for future flights. We do not recommend processing data directly from the SD card because SlantView will be limited by the speed of the USB or microsd port's connection. After selecting Organize data and navigating to the dataset on the SD card, selecting Copy data will create a local copy of the dataset at the directory location C:\SlantRange\Data[Grower][Field name]\crop_type] without deleting the dataset from the card. Selecting Move data will create a local copy 17

18 of the dataset at the directory location C:\SlantRange\Data[Grower][Field name]\crop_type] and delete the dataset from the card. Copy data and Move data are the equivalent of using copy and cut operations, respectively. If using a 1p, 2i, or 2p system Follow the four steps below to download data from your 1p, 2i, or 2p sensor. Connect the sensor (with attached AIS) to the computer via Ethernet cable, and boot up the sensor. Connect the sensor to SlantView Select a Dataset to download Clear data from the sensor Connect SlantView to the sensor When the sensor is connected to a computer via Ethernet cable, data can be downloaded and Processed in SlantView by selecting Sensor Download datasets as seen on the bottom of Figure 1, or by clicking the Connect to Sensor toolbar button boxed in red. Once connected, the window shown in Figure 2 will appear. Note The AIS does not need to be connected to the field sensor to download data. However, the lights on the AIS indicate when the sensor is ready to connect to SlantView, so it is advantageous to keep the AIS connected. As soon as the yellow light begins flashing on the AIS around seconds after bootup, press the button in the SlantView toolbar to connect. Figure 1: Loading data from your sensor Select a dataset to download 18

19 In the Dataset: drop down menu shown in Figure 2, select a particular Dataset or ALL, and enter a Grower and Field name. The text entered in the Grower and Field name boxes is used to name the folders created to store the downloaded data in your C:\SlantRange folder. If no Grower or Field name is entered, the data will be stored in folders named Unknown. If you are not sure what date-time stamp corresponds to the flight you are looking for, selecting Show coverage with a particular Dataset highlighted allows the user to view an outline of the flight. Figure 2: Processing data from your sensor Delete data from sensor The Delete from source check box allows you to clear data from the sensor as it is transferred to your computer. Your system will store approximately 3.5 hours of data before the AIS LED begins to blink red, indicating sensor memory is close to capacity. Approximately 1 hour after starting to blink red, the memory will be full. As a general rule, check the Delete from sensor option to make room for future data collection when downloading data. After clicking OK and waiting for the file transfer, the data can be accessed in the file explorer by selecting C:\SlantRange\Data\[Grower name]\[field name]. Folders will generally have several gigabytes of image data and will take a few minutes to load, progress is shown across the bottom of the Map Window. Note The SlantView window cannot be minimized during the downloading period, but other windows can be opened over the top of the Map Window so the user may continue working with other applications including other SlantView windows. Check to see if your computer's speed is limited: on many Windows computers, Balanced or Power Saver modes are default settings when running on battery power. Enabling your machine's version of High Performance mode will allow for much faster downloading and Processing. To check this setting, go to Control Panel Hardware and Sound Power Options as shown in Figure 3. 19

20 Figure 3: Windows high performance mode 20

21 Process Data Click here to see a quick tutorial video on how to Process your data! Process data that has been saved to your computer by following these four steps: Select a Dataset to Process Optional: select Store Locally, and enter a Grower and Field name to create a copy of the Dataset in your C:\SlantRange folder (if the data is already stored here, the Grower and Field names will be automatically populated) Select Processing type Review Processing settings Begin Processing Note Make sure your computer has at least as much available RAM as the size of the raw dataset folder you are processing (not including any previously created workspace files). Every hour of data collection is approximately 16 GB (standard consumer laptops have 8GB RAM). Depending on your operating system and other applications running in the background, as much as 40% of your total RAM may be unavailable to SlantView. If you are processing large datasets and notice dramatically slow computer performance or long processing times, open the Task Manager (CTRL+ALT+DEL, or CTRL+SHIFT+ESC) and make sure the memory usage is not at 100%. If the RAM usage is at or near 100%, the computer will resort to caching data on your hard drive (a process called "virtual memory", or a "paging file"). This will exponentially increase processing time from minutes to several hours, given the slower read and write speeds. If you plan on processing single flights of minutes we recommend upgrading to at least 16 GB of RAM. For individual datasets or combinations of datasets that sum to more than 45 minutes of flight time, we recommend at least 32 GB of RAM to consistently prevent use of paging files. Select dataset to Process To Process data stored on your hard drive, click Data Process dataset as seen in Figure 1 below. This will open a file explorer window to navigate to the folder containing the Dataset of interest. SlantView s file explorer window searches for folders not files; you will not see the individual.tif images within 21

22 the folder. To view all the images and files within a Dataset folder, open it in the traditional Windows file explorer. Figure 1: Select a Dataset for Processing If you move Datasets into new folders, be sure not to rename or delete the original folder name e.g. the T folder shown in Figure 1. Note For SlantView to open a Dataset, the folder name must be a 15 character time and date identifier in the form (year)(month)(day)t(hour)(minute)(seconds), matching the format of the folder name from initial download. Optional: enter a location to store the data Checking the Store locally box shown in Figure 2 will make a local copy of the Dataset in the C:\SlantRange\Data directory within folders named by the Grower and Field name you enter. Locally stored data (SATA connection) will load faster than data on an external drive which is limited by the speed of its USB connection. If disk space is a concern, use an external USB 3.0 drive or similar to preserve Processing speed and local hard drive space. If your data is stored outside the C:\SlantRange folder, and you do not want to create another copy of the data, leave the Store locally box unchecked, and leave the Grower and Field name blank. Note Loading data from cloud, local network, or other wireless storage will significantly increase Processing time; Ethernet, USB 3.0 and other wired connections are able to transfer data at a much higher rate. If you are not sure which field was flown in a particular Dataset, clicking Coverage may jog your memory by displaying outlines of the images and the decoded date-time stamp as shown in Figure 3. The blue box is a field boundary defined when processing a previous flight of this field (see the Trimming Content and Maps section of the View Data page for information on field boundaries). 22

23 Figure 3: Coverage quick-view Select Processing type Select a Process from the drop down menu shown in Figure 2: Figure 2: Processing Options Note If you have an active SlantView Basic or Pro license, but only see the options to Log metadata and/or Export images Log metadata The Log Metadata processing mode is used for troubleshooting, and when working with SLANTRANGE customer support. 23

24 Export images The Export images processing mode, available with SlantView Lite, Basic, and Pro subscriptions, exports all the calibrated imagery from the selected dataset(s) for use with third party software. Stress analysis The Stress analysis processing mode, available with SlantView Basic and Pro subscriptions, is for creating maps of intermediate and mature (closed row or canopy) crops. Use this mode for crops in more mature growth stages that do not meet the conditions needed for Population or Weed Detection. Weed detection The Weed detection processing mode, available with SlantView Pro subscriptions, is for identifying weeds in pre-emergence fields. In this mode, all vegetation detected in the field will be considered a weed. Note The Weed Detection Processing option is not to be confused with the Weed Detection Map. Weed Detection Processing will treat all vegetation as weeds. The Weed Detection Map that is part of Population Processing will identify weeds mixed in with crops. Population The Population processing mode, available with SlantView Pro Subscriptions; is for performing stand counts and weed detection in early stage crops. Population processing only works on early stage crops with bare soil exposed in between individual plants and rows. Generating accurate data products requires flight planning procedures that are unique to the current stage of the crops, as described in the Mission Planning section. In other words, a Dataset that was flown for crops in the Population stage will not Process as a Stress map, just as crops in a mature stage cannot be Processed into a Population map. See the View Data section for examples of Population versus Stress map products. Review Processing settings The following is a brief description of the settings that you may want to adjust in the various processing modes as part of a standard SlantView workflow. For a full description of each mode's settings, see the Adjusting Data Processing settings page of the section Advanced SlantView Functionality. Many of the settings apply to multiple Processing modes. Default settings for Exporting Images 24

25 If you are using SlantView Lite, or want to export your SLANTRANGE system's calibrated imagery to another software package, we recommend using the settings shown in Figure 3. Figure 3: Export images settings The Image exports options are 1. Convert pixel values to reflectance: depending on the intended use the exported images, you may wish to convert the pixels from false color to reflectance values. 2. Use JPEG compression: if working with a very large dataset and wanting to save memory space, you may want to select this option to JPEG compress the images. 3. Trim images to valid content: we recommend checking this box to remove pixels from the edges of each image plane that do not fall within the field of view of all the sensors. We recommend using the default settings in the Navigation and Sensor calibration sections. For more information on these settings, refer to the page Adjust Data Processing Settings in the Advanced SlantView Functionality section of the user guide. Default settings for Stress analysis processing Clicking the Settings button with Stress analysis selected opens the menu in Figure 4. 25

26 Figure 4: Default Stress analysis settings Starting from the top, the Sensor to airframe alignment offsets of Roll, Pitch, Heading: bias and Variable should be left at zero. These settings were created for the SLANTANGE 1P system; 2i, 2p, and 3p systems have inertial navigation and do not require offsets (see the page Manually Adjust Boresight Alignment in the Advanced SlantView Functionality section for more details). We recommend leaving Use compass data and Enhance geolocation checked. In some circumstances, SLANTRANGE technical support may recommend changing these settings and Reprocessing. Use the Altitude is constant option if mounting your sensor on a ground vehicle rather than an aircraft. We recommend leaving both Sensor Calibration settings checked: Auto boresight alignment and Auto flat field calibration. If flying a small field or short flight, less than ~200 images, where a comprehensive calibration profile cannot be reliably created, deselecting these settings to rely on factory calibration settings may improve results. In the Map Generation settings, select the maps you want to create: 1. NDVI Note If the Filtered box next to NDVI is checked, the Soil and Shadow Filters will be applied to the NDVI images, and the NDVI maps will show plant content only. Take a look at the Adjust Shadow and Soil Filters page in the Advanced SlantView Functionality section of this user guide for an example. The non-ndvi maps (Stress, Closure, Yield Potential, and Population) always use the Shadow and Soil Filters, and checking this box will not affect their appearance. 26

27 1. Stress 2. Vegetation fraction Measurement of canopy closure. 3. Yield Potential Combination of the stress and vegetation data products that predicts yield potential. Select an appropriate Map Resolution: The map resolution can be set between 30cm and 10m depending on the size of the plants in the field and the size of the field. Typically, larger fields (80 acres or more) are Processed at 30cm or 50cm. Higher resolution maps require slower airspeeds, lower altitudes, and thus more images, resulting in larger Datasets and increased Processing time. In most cases, selecting a higher map resolution does not provide any additional information. SlantView does not "throw away" any data when resolutions lower than the highest (30 cm) are selected. The resolution defines the size of the Map Window's colormap "tiles", essentially defining the number of pixels in a high resolution image (seen in the Image Window) that are binned together to create an individual "pixel" of the colormaps seen in the Map Window. Select your map resolution based on the spacing of crops in the field. As a general rule, pick a resolution that is larger than the width of the empty space between plants. For example, if your crops are spaced 1 meter apart and 50cm resolution is selected, the colormaps shown in the Map Window may have "tiles" where no pixels of plant content exist, resulting in a map with blank areas between rows of crops. If your crop spacing is 30cm and 1m map resolution is selected, one "tile" of the Map Window can include content measured from several plants, and the color of that tile in the Stress Map will be a more accurate representation the sum of the plant stress in that 1x1m area. For vineyards, orchards, and other crops, that have a relatively large amount of bare soil between the individual plants even when mature, feel free to Recalculate maps with a few different map resolutions above and below the row spacing. Maps showing empty space between rows or individual plants may provide a more intuitive representation of crop health in the field. The Image Window displays the same high resolution images regardless of the selected map resolution. If a field boundary was previously defined for the dataset, checking the Trim to field boundary if known option will automatically crop the map to that field boundary. Click OK to begin Processing! Default settings for Weed detection processing 27

28 Recall from Select Processing Type section above, that this mode should only be used for pre emergence fields, as all vegetation will be treated as weeds. Figure 5: Default Weed detection settings We recommend using the default settings within the Navigation and Sensor calibration sections. For more information on these settings, refer to the page Adjust Data Processing Settings in the Advanced SlantView Functionality section of the user guide. If a field boundary was previously defined for the dataset, checking the Trim to field boundary if known option will automatically crop the map to that field boundary. Default settings for Population processing Clicking the Settings button with the Population processing mode selected opens the menu in Figure 6. 28

29 Figure 6: Default settings for Population analysis 1. The Plant row spacing value must be entered to ensure accurate Population map statistics. If the crop is planted in skip rows, use the smallest distance between rows that exists throughout the field. If the Plant density is known, entering the values and selecting As Planted in the drop down menu in the bottom left will calibrate SlantView s color scaling such that green represents areas at the entered density, yellow and red represent areas populated less densely, and dark green represents areas populated more densely. 2. Select the Crop is planted in straight rows option if the crop is planted in linear, evenly spaced rows (e.g. uncheck this box if the crop is planted in skip rows). 3. The Population map that SlantView generates can be viewed at resolutions from 1m to 10 meters. As a general rule, pick a resolution that is larger than the width of the empty space between plants. (See the Choosing a resolution for Stress maps note above for further explanation.) 4. Selecting Computed will allow SlantView to automatically determine the Color Scale for the map, making green the 70th percentile of the plant density detected in the field. Note Setting the Plant row spacing is important for computing accurate Population map statistics, and should be set to the actual plant row spacing in the field. Plant density is for reference only, and can be left on the AUTO setting. Click OK and begin Processing! 29

30 View Data Click here to see a quick tutorial video on how to navigate the SlantView windows to view your data! Once a Dataset is Processed, SlantView's menus and toolbars offer a variety of options for gathering the information you need. The data viewing interface is made up of two main windows: the Map Window shown on the left of Figure 1, and the Image Window shown on the right of Figure 1. The Map Window contains most of Slantview s user controls, including controls for downloading data, Processing data, viewing data, editing data, and saving data. The Map Window shows the map products that are the sum of all the imagery from a given flight; the Image Window shows the high resolution image of the area in the map outlined by the white box following the cursor. When SlantView is first opened, the Map Window is active (the thin red border denotes the active window). Scroll your cursor through the Map Window to view high resolution images in the Image Window. The active window can be changed from the Map Window to the Image Window and back by double-clicking the left mouse button or pressing the Tab key. When control is passed from the Map Window to the Image Window, you can zoom, click, and drag the high resolution image. Figure 1: Map and Image Window The Map Window 30

31 The primary features of the Map Window discussed in this section are: The Title Bar The Menu The Toolbar The Map Window display The Title Bar The Title Bar at the very top of the Map Window displays the latitude and longitude of the cursor s location by default (pixel and UTM coordinates can be shown as well). The Stress, Vegetation fraction, Yield, or NDVI values at the cursor's current location are shown next to the coordinates depending on which map layer is active, Item A in Figure 2. The Menu The Menu includes the Data, Display, Settings, Sensor, and Help tabs, Item B in Figure 2 The Toolbar The Toolbar includes hotkeys for many common SlantView operations including opening and saving data and switching between map layers. Various map layers are available for viewing in the Map Window depending on which Processing mode was selected, and which maps were generated in the Settings menu. To switch between the layers, select a layer's icon from the toolbar. Item C in Figure 2. Figure 2: The Toolbar The Map Window display Basic information about the map is displayed in the upper left corner of the Map Window including the field name, the acreage of the map, the date and time the data was collected, and the area of a Custom Boundary within the map (if one has been drawn), Item D in Figure 2. To zoom in on the Map Window, scroll with your mouse wheel when the Map Window is active, or click on the plus and minus icons in the lower left corner of the window, Item E in Figure 2. 31

32 Figure 2: Map Window controls The Image Window The following primary features of the Image Window are discussed in this section. Viewing high resolution data layers Information displayed in the Title Bar Introduction to Filtering Viewing High Resolution Data Layers In addition to the color image that is shown in the high-resolution window by default, the same layers that are available in the Map Window are available in the Image Window. To change the layer that is displayed in the Image Window, press the number keys 0-9 on your keyboard when the Image Window is active. (1) GNDVI, (2) RNDVI, (3) ReNDVI, (4) Stress, (5) Color, (6) Raw Green Channel 1, (7) Raw Red Channel 2, (8) Raw Red Edge Channel 3, (9) Raw Near Infrared Channel 4. Information Displayed in the Title Bar The Title Bar at the top of the Image Window, Item A in Figure 3, displays the directory path to the data in the open Workspace. It also displays the name of the individual.tif image that is being displayed, and the data layer that is being displayed in the Image Window. 32

33 To zoom in on the image, scroll with your mouse wheel when the Image Window is active, or click the plus and minus buttons in the lower left corner, Item B in Figure 3. Figure 3: Image Window with various viewing options Color Left, Stress Middle, Green NDVI - Right The Map Products Example Population analysis for early stage crops Figure 4 is an example of a Population Dataset. Three maps are available for Datasets of crops in the Population stage: Weed Detections, Population Density, and Plant Size. Click the toolbar icons boxed in red to toggle between views. The Title Bar of the Map Window displays the plant density calculated for the active image. Zooming in on the Image Window, as seen on the right of Figure 4, shows weeds (orange) between rows of plants (green). Note The Weed Detection Map available in this Population dataset is not to be confused with the Weed Detection Processing type described on the Process Data page. If Weed Detection Processing was chosen for this Dataset instead of Population Processing, every plant detected would show up as a weed. Weed Detection Processing implies that the field is unplanted or the crops are pre-emergent. 33

34 Figure 4: Population analysis for early stage crops The Plant Size map is a relative measure of plant size (the area of each plant when viewed from above) to show where plants in a field are bigger and growing more quickly, vs smaller and getting a slower start. The colorbar on the left is set such that plants in the 75th percentile are assigned a size value of 1 and a color of green. From there a size of 0.8 means those plants are 80% as large as the plants in the 75th percentile, and 1.2 would be 120% as large as the plants in the green section. Example Stress analysis for mature crops Figure 5 is an example of a Stress Analysis Dataset gathered from a field of mature crops. The user is able view Green Normalized Differential Vegetation Index (Green NDVI), Red NDVI, Red Edge NDVI, Vegetation fraction, Stress, and Yield Potential if all are enabled in the settings menu. The SLANTRANGE Stress measurement is an absolute measurement that is a ratio of the four spectral bands measured by the sensor. It is our proprietary formula for identifying stress conditions with higher sensitivity and accuracy than can be achieved from industry standard two-band NDVI measurements. Given the variety of crops that our customers measure with their SLANTRANGE sensors, the system has to be adjustable to display a broad range of reflectivity profiles and stress conditions in an intuitive map. For example, say the sensor measures the reflectivity profile of healthy mature corn and computes stress values between.1 and.2. In another corn field, at an earlier growth stage, in a different part of the world with different soil, the sensor measures stress values between.4 and.5. In these extreme examples, the regions of corn with stress values of.1 and.4 are the healthiest areas from their respective fields. A particular field's conditions must be considered when comparing stress between different crops in different growth stages. At the moment, there is no library of multispectral reflectance signatures for stress conditions in every variant of every crop, in every type of soil, across the world, to reference when 34

35 assessing the level of stress and causes of stress, in your field. The Stress map shown in SlantView shows the range of Stress values across a field and makes it easy to isolate areas that may require immediate attention or action, but determining what the Stress values measured by the sensor actually correspond to (such as a specific pest, nutrient deficiency, disease etc.) is a question best answered by the agronomist or grower. The SLANTRANGE system can be referred to as a "rapid scouting tool", and the agronomist and grower are still an integral part of interpreting the data. We allow users to adjust the application of our Stress Color Scale to the actual Stress values measured by the sensor (this can cause confusion about the absolute nature of the stress measurement) to show the differences in the measured Stress values in an intuitive way. Typically, we adjust the color scale to show the least stressed areas in shades of green and the most stressed areas in shades of red. To easily compare one field to another, the user simply has to make sure that a particular shade of green corresponds to the same measured stress value in both SlantView Map Windows using the buttons next to the color scale to make incremental adjustments (e.g. setting the darkest shade of green to a stress value of.1). Typically, we adjust the Stress Color Scale to intuitively show the least stressed areas in shades of green and the most stressed areas in shades of red. The maps in the center, and right of Figure 5 show nonintuitively adjusted color scales. In center of Figure 5, the majority of the stress data (with measured stress values of.1 to.35) is mapped across shades of green. The areas of the field with the highest stress are shown in light green, and are difficult to distinguish from the healthy areas, also shown in green. Similarly, the example on the right has a color scale with the stressed and healthy areas mapped across shades of red. The well adjusted color scale of the map on the left intuitively shows the range of stress values measured in the field, with stressed areas clearly in yellow and red, and the healthy areas in green. Figure 5: Stress analysis for mature crops Annotating features 35

36 Features in the map can be annotated by following the three steps below, shown in Figure 6: With the Map Window active, move your cursor over the feature of interest Right-click and select Annotate feature Enter the annotation into the text box and click OK When you label a feature on the map using the annotation tool, the note will appear in the Map Window, as shown on the bottom of Figure 6, The Annotations will also be exported with the KMZ map file. To remove your annotations, select Display Annotations Clear, or to or hide them from view on the map, deselect the Show option. Figure 6: Annotating features Trimming content and maps Click here to see a quick tutorial video on how to crop your maps and create annotations! The Trim content tool allows the user to trim maps to a Custom Boundary. This tool can be used to eliminate data outside the boundary of the field. To create and trim a map to a Custom Boundary, follow these four steps: While active in the Map Window, right-click and select Custom boundary Polygon (or click the toolbar button boxed in red in the left panel of Figure 7) Left-click to select the edges of your polygon. 36

37 Right-click on the final point to finish the polygon. Right-click in the Map Window and select Trim content Maps only To custom boundary (Right panel of Figure 7). The data can be hidden from view, or eliminated from the Workspace entirely based on which Trim Content option is selected. The Trim content-maps only tool trims the map to the Custom Boundary, but all removed imagery remains with the Workspace and can be brought back upon clicking Recalculate Maps. Trim content-workspace trims the maps and removes all data outside the boundary from the Workspace. If content was trimmed from the Workspace, you will have to Reprocess from scratch to bring it back. To draw a rectangular or circular boundary quickly, right-click and select Custom boundary Rectangle or Circle. Figure 7: Custom map cropping If unsatisfied with the polygon, right click to complete the drawing then Right-click Custom boundary Destroy to erase and redraw. If you are viewing your cropped map and wish to return the map to its original size (and you did not crop the Workspace), click Recalculate maps on the far right of the toolbar, or select Settings Processing Recalculate maps. To trim a Workspace to a Custom Boundary: Right-click Trim content Workspace To custom boundary. If you trim a Workspace to a Custom Boundary, only the data within the boundary will be included when the Workspace is saved. This can be a handy tool if two fields have been flown in a single flight and the user wants to separate them into two unique Workspaces. Recalculating the maps or reopening the saved Workspace will not bring back the cropped content. However, the data is not removed from the original folder, so Reprocessing the data set from scratch will bring back any previously removed content. 37

38 To save a field boundary, Right-click Custom boundary Save. You will be prompted to enter a Grower and field name. The field boundary file is saved at C:\SlantRange\field definitions\[grower]\[field name]. A saved field boundary will be linked to the Dataset, and will show up whenver the Dataset is Reprocessed and whenever the Workspace is Recalculated. To view your saved custom boundaries, go to C:\SlantRange\field_definitions. SlantView also has the ability to highlight contours using the custom boundary tools. Adjust map transparency The transparency of the maps on top of the satellite imagery in the Map Window can be adjusted by holding the Alt key and clicking the up and down arrow keys. 38

39 Save Data Click here to see a quick tutorial video on exporting your data products! SlantView provides a number of options and formats for exporting data, shown in the Table below. Namely, the map layers, individual images or the entire set of images, which can be exported as KMZs, Shapefiles, and/or GeoTIFFs. With your mouse over the map right-click and select Export Maps Select to export either the current view or All views Select the file type for the exported map Data export options Map Views KMZ GeoTIFF SHP Map (current layer) X X Map (all layers) X X X Individual Image (current layer) X X Individual Image (all layers) X X All Images X 39

40 Figure 1: Map Window right-click menu Exporting the current view will save only the map layer that is currently displayed in SlantView. For example in Figure 1, the current view is the Stress layer, therefore selecting to export the current view will only save the Stress layer. Alternatively, saving all layers will save every map layer currently available in the Map Window, as chosen from the settings menu: NDVIs, Stress, Vegetation fraction, and/or Yield potential. The options for saving the data are KMZ (to view in Google Earth), Shapefile (for use with various Ag software packages), or multipage GeoTIFF (standard image format similar to an uncompressed jpeg). The exported file is saved within the folder that was initially selected to load the data. In addition to saving maps, the user may also save an individual image as KMZ or GeoTIFF by clicking Export Image to export the image currently in the Image Window. The right click options from the Map window only allow exporting all layers of an individual image, however a single layer KMZ image can be exported by making the image window active right-clicking and selecting Save image as KMZ Current view. 40

41 Note If the Store locally box was checked upon initial Processing, SlantView will save the exported files in C:\SlantRange\Data\[Grower]\[Field name] alongside all other files in the Dataset, otherwise the files will be saved in the folder that was selected to load the Dataset. If you are not sure where the files are being saved, the Image Window Title Bar displays the path to the active directory. Export Processed Images This performs the same function as selecting Export images in the Processing menu. View maps with Google Earth, and view map statistics From the Map Window, the user can save a KMZ archive and open the map products in Google Earth directly, as shown in Figure 2. The user can also view statistics in HTML format: Right click View maps with Google Earth. Right click View map statistics. 41

42 Figure 2: View maps with Google Earth, and view map statistics Statistics for each of the map products available in the Map Window, as selected in the Processing settings menu (NDVIs, Stress, Vegetation fraction, Yield potential, and any active Custom Filter maps), will be displayed in your default web browser, as shown in Figure 3. 42

43 Figure 3: HTML map statistics Saving a Workspace A Workspace can be saved in a variety of simple steps. The easiest of which is to select OK when prompted to save the Workspace upon opening a new Dataset or closing an active SlantView Map Window. Additionally, you can save the Workspace at any time by clicking: Data Save Workspace, as shown in Figure 4. Saving the Workspace preserves the most recent settings used to Process the data, as well as any Custom boundaries. Loading a Workspace allows the user to return to a map in a given desired state, instead of Reprocessing from scratch and reapplying settings. Also, a Workspace will open in a fraction of the time of Reprocessing an entire Dataset! 43

44 Figure 4: Saving a Workspace Note Workspaces are roughly the same size as the Dataset used to create them, therefore storing many Workspaces may take up signficant hard drive space. The Workspace file will be saved within the folder that was previously selected to load the Dataset. For Datasets loaded with the Store Locally or Download boxes checked, the Workspace will be at C:\SlantRange\Data\[Grower]\[Field name] alongside all other files in the Dataset. Saved Data Settings To adjust the settings of your saved data, open the Export Options menu from either the Data or Settings tabs in the main Map Window Menu, as shown in the Figure below: Figure 5: Navigating to the Export Options Menu From the menu in Figure 6, you can tailor your exports to various software packages: 44

45 Figure 6: The Export Options Menu For example, if uploading GeoTIFFs to Drone Deploy, select the Drone Deploy option. This coverts the images to reflectance data, and trims the images to valid content only. The trim option removes image content that is not shared between all four of the 2p and 3p's narrowband sensors in a given scene. To export processed images formatted for Pix4D, select the Pix4D option. This converts the images to reflectance data and does not trim to valid content. 45

46 Various softwares such as ArcGIS and Drone Deploy require different image settings for proper background transparency. Select the appropriate setting, Extra Plane or Extra Sample, depending on where you intend to export your maps (for GeoTIFFs and Shapefiles only). These settings are only allowed when exporting map layers Separately, rather than as Multipage TIFFs. An example of a Stress map GeoTIFF uploaded to Drone Deploy with the correct transparency setting: Extra Sample 46

47 An example of a Stress map GeoTIFF uploaded to Drone Deploy with the incorrect transparency setting: Extra Plane 47

48 48

49 Share Data This page introduces options for sharing map products with collaborators. If you are a subscriber to a cloud storage service (Dropbox, Google Drive, Microsoft OneDrive, etc.), one of the easiest ways to share map products with customers is to upload files to shared folders which can be set to sync automatically. Geotiffs, KMZs, and Shapefiles can be sent as attachments, requiring only a few megabytes per file depending on the size of the Dataset, as opposed to Workspaces which require a few gigabytes each. Please refer to the table on the Save Data page of this guide for a description of which map products can be saved in each file type. 1. Sharing Geotiffs A Geotiff is similar to an uncompressed JPEG, and can be opened with numerous traditional image viewing softwares that come pre-installed on standard consumer laptops: IrfanView, Windows Photo Viewer, Microsoft Paint, and Apple Photos, for example. They take a few seconds to download from a shared folder or attachment, and are the simplest way to share SlantView map products. All map products and all individual images from a Dataset can be exported as multi-page Geotiffs. As the name implies, Geotiffs include geographic metadata that allow platforms like Google Earth and Drone Deploy to overlay the maps with satellite imagery. GeoTIFFs do not include any annotations added to the maps in SlantView. 2. Sharing KMZs The KMZ is one of the primary file types used by Google Earth. Viewing a KMZ in Google Earth allows a user to see the SlantView map products and/or individual images on top of satellite imagery, similar to the view in the SlantView Map Window. KMZs include any annotations added to the maps in SlantView. 3. Sharing Shapefiles The Shapefile is a common file type used by farm managment software tools such as SMS, Farmers Edge, Granular, etc. 4. Sharing Workspaces Your coworkers and clients can download SlantView for free! By sharing Workspace files with your collaborators, you can work side by side in SlantView while viewing identical Map and Image Windows. Workspace files are an exact copy of all the data available in a given instance of SlantView, including all of your notes, edits, and settings modifications. The only SlantView functionality that is limited by loading a Workspace rather than Processing from scratch, is the ability 49

50 to modify initial Processing settings. All settings that are applied via Recalculating maps can be adjusted within a Workspace. From a Workspace, customers can adjust settings, export data products, and save Workspaces of their own. Workspaces include all the map products, images, and settings from a given instance of SlantView, and are therefore large files, typically a few gigabytes in size (approximately 1/2 the size of the full Dataset folder). If you have access to a cloud sharing platform with sufficient storage space, uploading and downloading Workspaces is the most comprehensive way to share data short of sharing the entire Dataset folder. 5. Sharing Dataset folders If your collaborators have downloaded their free version of SlantView, the most comprehensive way to share data is simply to share the entire Dataset folder. They can then Process from scratch in SlantView, and view any files you have saved within the Dataset directory. 6. Sharing via Drone Deploy Visit Drone Deploy's website for more information on the wide range of tools and capabilities they offer. Add SlantView map products to your Drone Deploy profile with a simple Geotiff upload! We are constantly exploring new platforms for sharing SLANTRANGE data, if there are any particular tools you, your coworkers, or clients would like to use, please reach out and let us know at slantrange.com, on or on Facebook! 50

51 Automated Statistics Reports The Statistics Report tools allow SLANTRANGE users to provide growers and collaborators with a summary of the conditions in the field as soon as Processing is complete, in convenient PDF or HTML formats. All map layers selected in the Processing Settings menu (that can be seen in the toolbar), including any custom filter map layers, will be shown in the PDF. Any Contours, Features, and Annotations will be overlaid on the maps. To generate the report, right click while active in the Map Window and select View Map Statistics, as shown in Figure 1. The PDF will open automatically in your default PDF reading application (if you have not selected a default application for reading PDFs, you may be prompted to do so). Figure 1: Generate a Map Statistics report The following images are from a sample Statistics report of a Stress Analysis flight: 51

52 Figure 2: Statistics Report Stress map 52

53 Figure 3: Statistics Report Vegetation Fraction map 53

54 Figure 4: Statistics Report Yield Potential map 54

55 Figure 5: Statistics Report Green NDVI map 55

56 Figure 6: Statistics Report Custom Area Density map High resolution images from the Image Window can also be added to the reports. While active in the Image Window right-click Save current image view--add to PDF report. Any image layer can be added (use number keys 1-9 to cycle through the layers: NDVIs, Stress, Yield potential, Vegetation fraction, and the four individual bands of the sensor) 56

57 Figure 7: High Resolution False-Color view (left) and Stress view (right) Note It is called "False-Color" (not true RGB) because the standard SLANTRANGE sensor band configuration inlcudes red and green, but not blue channels. The blue channel is interpolated for this view, which generally is a close match to what the human eye sees. When generating the PDF report, you will be prompted to add your own logo, as well as any specific notes. These are printed in the page header above the map on the left and right, respectively. 57

58 Data Uploader The SLANTRANGE team may ask for a dataset to help with troubleshooting. The data uploader application that installs with SlantView can be used to quickly share data. The application interface is simple, easy to use, and runs in the background. Open the SLANTRANGE Upload application. Note: If you just installed SlantView you will need to login first. The main application window is described below. To add data to the upload queue select one the SlantView data types. Select Add Workspace and navigate to the.workspace file to upload a SlantView workspace. Example: T workspace. Select Add Raw Image Upload and navigate to the data set folder name to upload all raw images and log files. Example: T Once the folder is selected, the information window will pop up. Enter in the information and be as specific and consistent as possible. 58

59 The Upload speed pull down allows the user to limit the upload bandwidth used. Select the Quit button when all data has been added to the queue to minimize the application window and keep uploading in the background. To check the progress of the uploader, click the SLANTRANGE tray icon on the bottom right of the desktop. Login to analytics.slantrange.com to view raw image uploads. 59

60 Advanced SlantView Functionality For the advanced user, a few tools to customize Datasets are discussed in this section: 1. Adjusting Shadow and Soil Filter thresholds 2. Creating Custom Filters 3. Assessing data quality: Displaying image footprints and setting metadata filters 4. Manually adjusting Boresight Alignment 5. Adjusting data processing settings 6. Measuring NDVI and Reflectance 7. Combining multiple Datasets into one map 8. Understanding file formats for integration with 3rd party software. 9. Batch Processing data 10. Updating sensor firmware 60

61 Adjust Shadow and Soil Filtering Settings Introduction to Filtering The following animations show how SlantView is capable of simultaneously highlighting and removing non-crop content including shadows between rows, areas of bare soil or road, and areas of different plants from its data products. When active in the Image Window, press the number keys 1, 2, and 3 to show the Green, Red, and Red Edge NDVI high resolution views. In the Settings Processing menu, checking the Filtered box next to the NDVI map product option will remove non-crop content and display NDVI maps similar to those in Figure 1. Leaving the Filtered box unchecked will display NDVI views as in Figure 2, where non-crop content is not blacked out. Figure 1: green (left), red (center), and red edge (right) filtered NDVI views 61

62 Figure 2: green (left), red (center), and red edge (right) unfiltered NDVI views Figure 3: Unfiltered high-resolution Stress image 62

63 Figure 4: Filtered high-resolution Stress image Figures 3 and 4 show a high resolution Stress image filtered and unfiltered. In Figure 4, the weeds, shadows, and bare soil are blacked out, while the stressed areas of crops show up as the stressed content highlighted in red. The user typically does not want bare soil, weeds, and other non-crop content displayed in the maps, and therefore has the option to create filtered or unfiltered Stress and NDVI maps. If the goal is to measure the health of the crops only, SlantView can filter out the weeds (Figure 4 item A), shadows (Figure 4 item B), and bare soil (Figure 4 item C). The process of adjusting Shadow and Soil Filters is described in the following section. To compute meaningful statistics on plant health, plant material must be separated from soil and shadows, as shown in Figure 5. If the default filtering (recall the Soil Filter threshold set to 0.5 in the **Process Data"" section) leaves too much soil or too little plant material, manually adjusting the thresholds may improve results. 63

64 Figure 5: Plant and soil reflectance filtering To make precise adjustments to soil and shadow thresholds with visual feedback: 1. Press the F7 key when the Image Window is active. The Shadow and Soil Filter settings will be displayed at the top of the Image Window, shown boxed in red in Figure (Optional) Bring both filters down to zero to see the exact content each filter identifies as the strength is increased. 3. Increase the Shadow Filter until all shadows have been greyed out, by holding down the Ctrl key while scrolling the mouse wheel, or use the on-screen slider bar. 4. Increase the Soil Filter until all non-crop content has been greyed out, by holding down the Ctrl + Shift keys while scrolling the mouse wheel, or use the on-screen slider bar. 5. Move back to the Map Window by double-clicking or pressing the Tab key, and check a few images across the map with the new filter thresholds before Recalculating the maps. 6. Recalculate the map by clicking the Recalculate maps button on the lower right side of the settings window or the far right button on the toolbar. The Shadow and Soil Filters in the Image Window in Figure 6 (right side), have been zeroed out. 64

65 Figure 6: Example of no Shadow or Soil Filter Figure 7: Left: Shadow Filtering only, Right: combination Shadow and Soil Filtering If non-plant material accounts for most of the red areas of your Stress Map, manually adjusted filters may be more effective at highlighting potential problem areas within a field. After visually adjusting Soil and Shadow Filters to separate plant material from soil, roadways, equipment, and other non-plant material, Recalculate maps based on your new thresholds. 65

66 Using Custom filters and Smart Detection In brief, Custom Filters aka Smart Detection allow the user to find content that matches a specific spectral signature, and they are useful for both highlighting and removing content from your maps. To create a map of areas of a field that match a designated spectral signature, follow these seven steps: 1. Set image orientation to As Collected 2. Identify an image with your feature of interest 3. Activate the Image Window and press F7 to adjust the filter settings 4. Identify an area of interest in the Image Window by holding CTRL, then left clicking and dragging your mouse to create a rectangle around the content 5. Create an Area Density or Stress Map, or filter out content by Absolute or Relative spectral relation to your highlighted region 6. Adjust the strength of the new Custom Filter 7. Recalculate maps to include the new Custom Filter The following Custom Filter demonstration is an example of creating an Area Density map of dodder weed in an alfalfa field: Set image orientation to As Collected Begin by setting the image orientation to as collected, by clicking Display Image View Orientation As collected, as shown in Figure 1. The default Image Window view is set as North-up to align the images with the Map Window view which is always oriented North-up. When the Image view is changed from North-up to As collected, images will be rotated such that the top of the image in the Image Window faces the front of the aircraft at the time the image was captured. For example, if the flight legs were flown East-West, the images taking while flying East to West will be rotated 90 degrees counterclockwise relative to the North-up view, and the images flown West to East will be rotated 90 degrees clockwise relative to North up. Note 66

67 Image orientation will not change until a new image footprint is highlighted in the Map Window. If the orientation was changed while active in the Image Window (look for the red border), double-click or press the Tab key to activate the Map Window and move your cursor to a new image footprint. The Image Window will update the orientation, and you can proceed with creating the Custom Filter. Figure 1: Changing image orientation Identify an image with your feature of interest Double-click on an image footprint in the Map Window. In the Image Window, zoom in on the feature and press the F7 key. Adjust Shadow and Soil Filters if the content you intend to select is already partially filtered out. A Custom Filter cannot be created if the Custom Filter Region is already filtered out by 67

68 Shadow and Soil Filters. It is easiest to create a Custom Filter and vary its strength while the Shadow and Soil Filters are set to zero. You can always return the Shadow and Soil Filters to their previous values once the Custom Filter is set. Create the Custom Filter Region While holding down the control key, left-click and hold the mouse button to draw a rectangular box around the feature of interest. Figure 2 shows a box drawn over a patch of dodder weed in alfalfa. Figure 2: Drawing a Custom Filter Region Create an area density map SlantView isolates points throughout the Dataset with reflectance spectra matching that of the highlighted box. To create a map of all these areas, right click inside the box and select Create map Area Density, as shown in Figure 3. 68

69 Figure 3: Creating an area density map from a Custom Filter Similar to the Soil and Shadow Filters, the image content identified by the Custom Filter Region is covered with solid grey pixels. Note If Soil, Shadow, or other Custom Filters are active, there may be several grey-ish colors overlaid on your image. If you cannot see the Soil, Shadow, or Custom Filtered pixels, press the F7 key to activate the Filter View in the Image Window. Adjust the strength of the Custom Filter To adjust the strength of the Custom Filter, bring your cursor inside the box and hold the control key while scrolling up or down with your mouse wheel. Shadow and Soil Filters can be adjusted alongside the Custom Filter by bringing the cursor outside the Custom Filter Region and holding Ctrl or Ctrl+Shift keys and scrolling the mouse wheel, as described in section Adjusting Shadow and Soil Filtering Settings. The image on the left of Figure 4 shows a weaker Custom Filter; the image on the right shows a stronger Custom Filter that has identified more pixels, and effectively highlighted all the dodder weed in the image. Zoom out and move around the images in the Image Window when increasing and decreasing the strength of the Custom Filter to see how other areas are affected. 69

70 Figure 4: Adjusting strength of a Custom Filter After setting an appropriate Custom Filter strength, return the Shadow and Soil Filters to their previous levels, and click the Recalculate maps button on the far right of the toolbar, or in the settings menu to generate the Area density map. The new map layer will be named Custom-Density by default. The Custom Filter map will show up as a new tab at the bottom of the list in the top right of the Map Window, seen in Figure 5. Note To edit a Custom filter, you must navigate back to the red box of the Custom Filter Region where it was created. To save time, consider leaving an annotation on the image, explained in the section View Data. Alternatively, if you need to delete an old Custom Filter that you are having trouble finding, simply draw a new box anywhere, right click and select Delete all filters, then Recalculate the maps. 70

71 Figure 5: Custom Filter area density map Four different filtering options including the Area Density map can be used in SlantView: 1. Filter out Absolute An absolute Custom Filter removes the highlighted pixels from the map products in the same way as Shadow and Soil Filters. 2. Filter out Relative The Relative filter functions the same way as the Absolute filter, but uses normalized rather than absolute reflectance data. In general, this mode is useful in fields with different levels of illumination, such as tall crops (e.g. vineyards and orchards) that cast shadows on the data you want to filter out. 3. Create Custom Stress Map After drawing a Custom Filter Region, right click Create map Stress. If using the Custom Filter to highlight plant content rather than content to be removed (soil, weed, etc.), this option will make a Stress map that includes only the content that was highlighted by the Custom Filter. 4. Create Custom Area Density Map As seen in the tutorial above, the Area density Custom Filter creates a map matching what is highlighted in the Custom Filter Region. Note The Soil and Shadow Filters, and the Filter out and Create map Custom Filters are NOT mutually exclusive. The Filter out Relative or Absolute filters can include the same content as custom Area density and Stress maps, and Soil and Shadow Filters. 71

72 72

73 Assess Data Quality The quality of your data is dependent on the quality of your mission planning. It is imperative that you read and understand the content on the 2p or 3p Flight Checklist page and in the Mission Planning section, regarding sensor mounting and aircraft flight planning. Several tools in SlantView are built in to help you visualize your flights and recognize factors that affect data collection. Displaying image footprints Aircraft heading, roll, pitch, yaw, and speed over ground all affect the alignment and overlap of images relative to one another. Click Display Images footprints to show image outlines overlaid on the field in the Map Window. For best results, fly in straight flight legs with consistent velocity and orientation. To produce high quality maps, images should have at least 30% along-track overlap and 20% sidelap, as shown in Figure 1. 73

74 Figure 1: Image along track overlap and sidelap To get a sense of how aircraft orientation affects along-track overlap, consider a multirotor at the beginning of a new flight leg. After the multirotor turns 180 degrees at the edge of a field, it will accelerate to the velocity set in the mission planning software for its next pass. This acceleration causes the multirotor to pitch forward, angling the field senor slightly behind the aircraft. As it reaches its max velocity, the acceleration goes to zero, and the field sensor returns to nadir orientation (as established by the field sensor s angled mounting bracket). Along-track image overlap will vary according to multirotor acceleration throughout a given flight; we recommend flying slightly beyond the edges of the field to allow the multirotor to change direction and accelerate while it is not imaging content of interest. SlantView filters out images captured during turns using the Set Filters parameters in the Settings Processing menu based on thresholds for heading, roll, pitch, and yaw. To get a sense of how aircraft orientation affects image sidelap, consider an aircraft in the presence of a crosswind. A multirotor s roll angle will change to maintain the given heading, tilting the field sensor to the port or starboard side. A fixed wing aircraft s yaw angle will change in the presence of a crosswind, and the aircraft will crab to maintain a given heading, resulting in image footprints that are slighlty rotated 74

75 relative to the heading of the given flight leg. Image rotation can compromise image overlap (as seen in Figure 3). For best results, fly perpendicular to the crosswind. On a multirotor, the roll angle will be mirrored from one flight leg to the next, and even though the field sensor is leaning to the port or starboard side, image overlap from one pass to the next will be consistent. When flying parallel to the wind, the pitch angle is dramatically higher when flying into the wind versus with the wind, comprising consistent along-track image overlap, potentially reducing data quality. Keep the relationships between weather conditions, aircraft orientation, and resulting image location in mind when planning missions, and diagnosing map artifacts. Lack of overlap can result in poor image registration and holes in your maps. For reference, two maps image footprints collected from high quality mission plans are shown below. Notice the consistent along-track image overlap and image sidelap, combined with nearly perfect parallel flight legs in Figure 2. Better flights produce better data. However, you do not need perfect weather conditions or perfect flight planning to produce high quality data. Notice the changes in aircraft heading present in the flight legs of Figure 3; flown with slightly higher overlap to compensate for crosswind, SlantView is still able to produce very high quality maps. The calibrated SLANTRANGE sensor combined with SlantView s processing and filtering techniques have extracted high quality results around the world in all kinds of weather, in all stages of growth. 75

76 Figure 2: Very high quality flight, consistent image overlap 76

77 Figure 3: High quality flight, maintaining overlap in the presence of crosswind Displaying image metadata Click Display Image view Show metadata to view the aircraft roll, pitch, yaw, heading, altitude, and speed over ground statistics shown in Figure 4. After enabling the metadata view, while active in the Map Window, drag the cursor over an image to see its metadata displayed in the top left of the Image Window. Combined with viewing Image Footprints, image metadata will help you identify sources of error. 77

78 Figure 4: Image metadata Identify a few flight legs and move the mouse from one image to the next in order with the image numbers shown in the Title Bar (boxed in red in Figure 4). Assess the variation in roll, pitch, yaw, heading, altitude, and speed over ground across the flight leg. Good flight legs will show minimal variation in these values. Large differences of about 5-10º in a given flight leg, could be the result of gusting winds, aircraft instability, etc. If there are gaps in flight legs that look like missing images, it is possible that SlantView excluded images from the map based on the images' metadata as established by the Set Filters thresholds in the Settings menu. Setting metadata filters In the Settings menu, edit the Set filters parameters to display images within a given percentage of steady-state altitude, pitch, roll, yaw, or heading. The number of images used in a given map based on the filter settings is shown boxed in red in Figure 5. To remove more or fewer images from the map, the user can increase or decrease the width of these filters on aircraft attitude. Images can be removed from the Workspace by narrowing the filters and Recalculating. However, to increase the width of the filter, you must Reprocess a Dataset from scratch with the wider tolerances indicated in the Settings Set filters menu before pressing OK to Process the Dataset. Note If the default SlantView filters throw out too many images in a given Dataset, improper mission planning or poor flight conditions are almost certainly limiting the quality of your data. 78

79 Figure 5: Aircraft attitude filtering options Note Do not attempt to manually remove images from a Dataset folder when viewed in the file explorer. SlantView uses the metadata collected from all images taken from takeoff through landing (even images excluded from map generation) to establish ground elevation, flight altitude, flight leg direction, and a multitude of other critical measurements. Removing images can compromise the integrity of the flight metadata. SlantView will automatically exclude unnecessary images from your maps. 79

80 Adjust Data Processing Settings Factory calibration and flight-specific automatic calibration methods are used by your SLANTRANGE system to produce clear, repeatable data across a wide variety of conditions and crops. We are confident that the default settings described on the Process Data page will produce high quality results for the majority of users. However, every mission and every field is different, and SLANTRANGE systems are constantly being tested on new aircraft and new crops. In some cases, default algorithms may not produce the best results. The following settings are intended to give users backups and workarounds to improve data quality. 80

81 Settings for Exporting Images If you are using SlantView Lite, or want to export your SLANTRANGE system's calibrated imagery to another software package, use the Export images processing option. The default settings are shown below: Figure 3: Export images settings Navigation settings: Use compass data What is it? If your compass calibration, as performed with the 3p Toolbox app, is not up-to-date with your location and/or aircraft and mounting setup, or the compass data has significant interference in flight, changing this setting may improve map products. When should I use it? 81

82 If you have turned on image footprints and see clear misalignment between flight legs, processing with Use compass data disabled may improve results. Enhance geolocation What is it? This setting applies additional processing to the sensor's attitude metadata. When should I use it? We recommend always leaving this setting on for initial processing. If you enable Image footprints and see rotational misalignment or image placement errors, processing with Enhance geolocation disabled may improve results. Altitude is constant What is it? This setting is for users mounting specialized SLANTRANGE systems to ground vehicles. When should I use it? (Do not mount your SLANTRANGE sensor to a ground vehicle unless it was specially manufactured for this use case) Only enable this setting if your sensor is mounted to a ground vehicle. If this option is selected, you must manually enter the sensor's altitude above ground. Sensor Calibration settings Auto boresight alignment What is it? This setting applies flight-specific calibration to the alignment of each of the system's narrowband sensors. When should I use it? We recommend always leaving this setting enabled upon initial processing. If flying at very low altitude (approximately 15 meters or less), disabling this setting may improve results. 82

83 Auto flat field calibration What is it? This setting applies flight-specific calibration to vignetting correction processing. When should I use it? We recommend always leaving this setting enabled upon initial processing. If processing a very short flight, disabling this setting may improve results. Image export options 1. Convert pixel values to reflectance Depending on the intended use the exported images, you may wish to convert the pixels from false color to reflectance values. 2. Use JPEG compression If working with a very large dataset and wanting to save memory space, you may want to select this option to JPEG compress the images. 3. Trim images to valid content We recommend checking this box to remove pixels from the edges of each image plane that do not fall within the field of view of all the sensors. 83

84 Settings for Stress analysis processing Clicking the Settings button with Stress analysis selected opens the menu in Figure 4. Figure 4: Default Stress analysis settings Navigation settings: Sensor to airframe alignment, degrees What is it? These settings are intended for use with the SLANTRANGE 1p system which does not have inertial measurement. 2i, 2p, and 3p systems have inertial measurement and do not require offsets, however in rare cases of improper mounting and IMS interference, these settings allow some freedom to improve results. When should I use it? Unless you have already processed the dataset and used the Boresight alignment tool to find the proper offsets, leave these settings at 0 upon processing. Refer to the page Manually Adjust Boresight Alignment for instructions on how to apply these settings from within a Workspace. 84

85 Use compass data What is it? If your compass calibration, as performed with the 3p Toolbox app, is not up to date with your location and aircraft/mounting setup, or the compass data has significant interference in flight, changing this setting may improve map products. When should I use it? If you have turned on image footprints and see clear misalignment between flight legs, processing with Use compass data disabled may improve results. Enhance geolocation What is it? This setting applies additional processing to the sensor's attitude metadata. When should I use it? We recommend always leaving this setting on for initial processing. If you enable Image footprints and see rotational misalignment or image placement errors, processing with Enhance geolocation disabled may improve results. Altitude is constant What is it? This setting is for users mounting specialized SLANTRANGE systems to ground vehicles. When should I use it? (Do not mount your SLANTRANGE sensor to a ground vehicle unless it was specially manufactured for this use case.) Only enable this setting if your sensor is mounted to a ground vehicle. If this option is selected, you must manually enter the sensor's altitude above ground. Sensor Calibration settings Auto boresight alignment 85

86 What is it? This setting applies flight-specific calibration to the alignment of each of the system's narrowband sensors. When should I use it? We recommend always leaving this setting enabled upon initial processing. If flying at very low altitude (approximately 15 meters or less), disabling this setting may improve results. Auto flat field calibration What is it? This setting applies flight-specific calibration to vignetting correction processing. When should I use it? We recommend always leaving this setting enabled upon initial processing. If processing a very short flight, disabling this setting may improve results. Map Generation settings Select the maps you want to create: NDVI If the Filtered box next to NDVI is checked, the Soil and Shadow Filters will be applied to the NDVI images, and the NDVI maps will show plant content only. Take a look at the Adjust Shadow and Soil Filters page in the Advanced SlantView Functionality section of this user guide for an example. The non-ndvi maps (Stress, Closure, Yield Potential, and Population) always use the Shadow and Soil Filters, and checking this box will not affect their appearance. Stress Vegetation fraction Yield Potential Combination of the stress and closure data products to predict yield. If a field boundary was previously saved for the dataset, selecting Trim to field boundary will automatically crop the dataset to that field boundary. To see if a field boundary exists for the data, open the Coverage map before processing. 86

87 Select an appropriate Map resolution: The map resolution can be set between 30cm and 10m depending on the size of the plants in the field and the size of the field. Typically, larger fields (80 acres or more) are Processed at 30cm or 50cm. Higher resolution maps require slower airspeeds, lower altitudes, and thus more images, resulting in larger Datasets and increased Processing time. In most cases, selecting a higher map resolution does not provide any additional information. SlantView does not "throw away" any data when resolutions lower than the highest (30 cm) are selected. The resolution defines the size of the Map Window's colormap "tiles", essentially defining the number of pixels in a high resolution image (seen in the Image Window) that are binned together to create an individual "pixel" of the colormaps seen in the Map Window. Select your map resolution based on the spacing of crops in the field. As a general rule, pick a resolution that is larger than the width of the empty space between plants. For example, if your crops are spaced 1 meter apart and 50cm resolution is selected, the colormaps shown in the Map Window may have "tiles" where no pixels of plant content exist, resulting in a map with blank areas between rows of crops. If your crop spacing is 30cm and 1m map resolution is selected, one "tile" of the Map Window can include content measured from several plants, and the color of that tile in the Stress Map will be a more accurate representation the sum of the plant stress in that 1x1m area. For vineyards, orchards, and other crops, that have a relatively large amount of bare soil between the individual plants even when mature, feel free to Recalculate maps with a few different map resolutions above and below the row spacing. Maps showing empty space between rows or individual plants may provide a more intuitive representation of crop health in the field. The Image Window displays the same high resolution images regardless of the selected map resolution. 87

88 Settings for Weed detection processing Recall from Select Processing Type section above, that this mode should only be used for pre emergence fields, as all vegetation will be treated as weeds. Figure 5: Default Weed detection settings Navigation settings: Sensor to airframe alignment, degrees What is it? These settings are intended for use with the SLANTRANGE 1p system which does not have inertial measurement. 2i, 2p, and 3p systems have inertial measurement and do not require offsets, however in rare cases of improper mounting and IMS interference, these settings allow some freedom to improve results. When should I use it? Unless you have already processed the dataset and used the Boresight alignment tool to find the proper offsets, leave these settings at 0 upon processing. Refer to the page Manually Adjust Boresight Alignment for instructions on how to apply these settings from within a Workspace. 88

89 Use compass data What is it? If your compass calibration, as performed with the 3p Toolbox app, is not up to date with your location and aircraft/mounting setup, or the compass data has significant interference in flight, changing this setting may improve map products. When should I use it? If you have turned on image footprints and see clear misalignment between flight legs, processing with Use compass data disabled may improve results. Enhance geolocation What is it? This setting applies additional processing to the sensor's attitude metadata. When should I use it? We recommend always leaving this setting on for initial processing. If you enable Image footprints and see rotational misalignment or image placement errors, processing with Enhance geolocation disabled may improve results. Altitude is constant What is it? This setting is for users mounting specialized SLANTRANGE systems to ground vehicles. When should I use it? (Do not mount your SLANTRANGE sensor to a ground vehicle unless it was specially manufactured for this use case.) Only enable this setting if your sensor is mounted to a ground vehicle. If this option is selected, you must manually enter the sensor's altitude above ground. Sensor Calibration settings Auto boresight alignment What is it? 89

90 This setting applies flight-specific calibration to the alignment of each of the system's narrowband sensors. When should I use it? We recommend always leaving this setting enabled upon initial processing. If flying at very low altitude (approximately 15 meters or less), disabling this setting may improve results. Auto flat field calibration What is it? This setting applies flight-specific calibration to vignetting correction processing. When should I use it? We recommend always leaving this setting enabled upon initial processing. If processing a very short flight, disabling this setting may improve results. Map Generation settings If a field boundary was previously saved for the dataset, selecting Trim to field boundary will automatically crop the dataset to that field boundary. To see if a field boundary exists for the data, open the Coverage map before processing. Select an appropriate Map Resolution: The map resolution can be set between 30cm and 10m depending on the size of the plants in the field and the size of the field. Typically, larger fields (80 acres or more) are Processed at 30cm or 50cm. Higher resolution maps require slower airspeeds, lower altitudes, and thus more images, resulting in larger Datasets and increased Processing time. In most cases, selecting a higher map resolution does not provide any additional information. SlantView does not "throw away" any data when resolutions lower than the highest (30 cm) are selected. The resolution defines the size of the Map Window's colormap "tiles", essentially defining the number of pixels in a high resolution image (seen in the Image Window) that are binned together to create an individual "pixel" of the colormaps seen in the Map Window. 90

91 Select your map resolution based on the spacing of crops in the field. As a general rule, pick a resolution that is larger than the width of the empty space between plants. For example, if your crops are spaced 1 meter apart and 50cm resolution is selected, the colormaps shown in the Map Window may have "tiles" where no pixels of plant content exist, resulting in a map with blank areas between rows of crops. If your crop spacing is 30cm and 1m map resolution is selected, one "tile" of the Map Window can include content measured from several plants, and the color of that tile in the Stress Map will be a more accurate representation the sum of the plant stress in that 1x1m area. For vineyards, orchards, and other crops, that have a relatively large amount of bare soil between the individual plants even when mature, feel free to Recalculate maps with a few different map resolutions above and below the row spacing. Maps showing empty space between rows or individual plants may provide a more intuitive representation of crop health in the field. The Image Window displays the same high resolution images regardless of the selected map resolution. 91

92 Settings for Population processing Clicking the Settings button with the Population processing mode selected opens the menu in Figure 6. Figure 6: Default settings for Population analysis Navigation settings: Sensor to airframe alignment, degrees What is it? These settings are intended for use with the SLANTRANGE 1p system which does not have inertial measurement. 2i, 2p, and 3p systems have inertial measurement and do not require offsets, however in rare cases of improper mounting and IMS interference, these settings allow some freedom to improve results. When should I use it? Unless you have already processed the dataset and used the Boresight alignment tool to find the proper offsets, leave these settings at 0 upon processing. Refer to the page Manually Adjust Boresight Alignment for instructions on how to apply these settings from within a Workspace. Use compass data 92

93 What is it? If your compass calibration, as performed with the 3p Toolbox app, is not up to date with your location and aircraft/mounting setup, or the compass data has significant interference in flight, changing this setting may improve map products. When should I use it? If you have turned on image footprints and see clear misalignment between flight legs, processing with Use compass data disabled may improve results. Enhance geolocation What is it? This setting applies additional processing to the sensor's attitude metadata. When should I use it? We recommend always leaving this setting on for initial processing. If you enable Image footprints and see rotational misalignment or image placement errors, processing with Enhance geolocation disabled may improve results. Altitude is constant What is it? This setting is for users mounting specialized SLANTRANGE systems to ground vehicles. When should I use it? (Do not mount your SLANTRANGE sensor to a ground vehicle unless it was specially manufactured for this use case.) Only enable this setting if your sensor is mounted to a ground vehicle. If this option is selected, you must manually enter the sensor's altitude above ground. Sensor Calibration settings Auto boresight alignment What is it? 93

94 This setting applies flight-specific calibration to the alignment of each of the system's narrowband sensors. When should I use it? We recommend always leaving this setting enabled upon initial processing. If flying at very low altitude (approximately 15 meters or less), disabling this setting may improve results. Auto flat field calibration What is it? This setting applies flight-specific calibration to vignetting correction processing. When should I use it? We recommend always leaving this setting enabled upon initial processing. If processing a very short flight, disabling this setting may improve results. Crop Details The Plant row spacing value must be entered to ensure accurate Population map statistics. If the crop is planted in skip rows, use the smallest distance between rows that exists throughout the field. Select the Crop is planted in straight rows option if the crop is planted in linear, evenly spaced rows (e.g. uncheck this box if the crop is planted in skip rows). Map Generation settings If the Plant density is known, entering the values and selecting As Planted in the drop down menu in the bottom left will calibrate SlantView s color scaling such that green represents areas at the entered density, yellow and red represent areas populated less densely, and dark green represents areas populated more densely. Selecting Computed will allow SlantView to automatically determine the Color Scale for the map, making green the 70th percentile of the plant density detected in the field. If a field boundary was previously saved for the dataset, selecting Trim to field boundary will automatically crop the dataset to that field boundary. To see if a field boundary exists for the data, open the Coverage map before processing. The Weed map and Plant size map are additional map layers you can choose to generate along with the Population density layer which is always shown. 94

95 Setting the Plant row spacing is important for computing accurate Population map statistics, and should be set to the actual plant row spacing in the field. Plant density is for reference only, and can be left on the AUTO setting. Select an appropriate Map Resolution: The map resolution can be set between 30cm and 10m depending on the size of the plants in the field and the size of the field. Typically, larger fields (80 acres or more) are Processed at 30cm or 50cm. Higher resolution maps require slower airspeeds, lower altitudes, and thus more images, resulting in larger Datasets and increased Processing time. In most cases, selecting a higher map resolution does not provide any additional information. SlantView does not "throw away" any data when resolutions lower than the highest (30 cm) are selected. The resolution defines the size of the Map Window's colormap "tiles", essentially defining the number of pixels in a high resolution image (seen in the Image Window) that are binned together to create an individual "pixel" of the colormaps seen in the Map Window. Select your map resolution based on the spacing of crops in the field. As a general rule, pick a resolution that is larger than the width of the empty space between plants. For example, if your crops are spaced 1 meter apart and 50cm resolution is selected, the colormaps shown in the Map Window may have "tiles" where no pixels of plant content exist, resulting in a map with blank areas between rows of crops. If your crop spacing is 30cm and 1m map resolution is selected, one "tile" of the Map Window can include content measured from several plants, and the color of that tile in the Stress Map will be a more accurate representation the sum of the plant stress in that 1x1m area. For vineyards, orchards, and other crops, that have a relatively large amount of bare soil between the individual plants even when mature, feel free to Recalculate maps with a few different map resolutions above and below the row spacing. Maps showing empty space between rows or individual plants may provide a more intuitive representation of crop health in the field. The Image Window displays the same high resolution images regardless of the selected map resolution. For more details, see the Map Resolution Discussion page of this User Guide. 95

96 File Formatting 1. GeoTIFF: GeoTIFF stored as uint8 values. 2. KMZ: Google Earth file containing the data layers selected from the Processing settings menu. 3. Shapefile:.shp,.dbf,.shx, shapefiles containing the data layers selected from the Processing settings menu. Population Analysis Map Products 1. Plant Population Density: The population value = 1.5 x the uint8 value (0-255) x the value coded as a float in the 2nd 4 bytes of TIFFTAG_SR_MAPSCALE (65005); and, where the units of the population are coded in the 3rd 4 bytes of TIFFTAG_SR_MAPSCALE (65005) as: 4 = per sq meter, 5 = per acre, 6 = per hectare, 7 = per sq km, 8 = per sq statute mile. 1. Plant Size Distribution: uint8 space scales to , where 1.0 is the 75th percentile of the median plant size. Stress Analysis Map Products This is a 345-acre field of early season Brazilian corn at approximately V5-V6. One of our customers sampled an 90-acre section in the southern quarter of the field. While NDVI and color imagery shows this to be a field of fairly uniform condition, the vegetation fraction measurement shows substantial variability in biomass which dominates the yield potential data product. The 6-page GeoTIFF saved for Stress Analysis: 1. Vegetation Stress uint8 space scales to relative stress condition (proprietary formula) 2. Vegetation Fraction uint8 space scales to 0-100% canopy closure 3. Yield Potential uint8 space scales to 0-100% relative potential (proprietary formula) 4. Green NDVI ( ) / ( ), uint8 space scales to

97 5. Red NDVI ( ) / ( ), uint8 space scales to Red-edge NDVI ( ) / ( ), uint8 space scales to If applicable: Custom area density, uint8 space scales to 0-100% of ground covered by signal type Shapefile formatting SlantView exports two kinds of shapefiles, both polygon type. One is the contour or custom boundary created in map window - which can be exported into a shapefile. Another is the map export into shapefile: When a map is selected for export into a shapefile, it is broken down into square boxes (5x5 m, IIRC). Each box is exported as a separate polygon, with four Lon/Lat coordinates of the corners. If a single map is exported into a shapefile, the database file will contain the name of the values for each polygon and the values themselves (mean value for the 5x5 meter area). If multiple maps are exported into a shapefile, the database file will contain the values (5x5m block averages) and names of all map entries. Output format: four files with the same prefix are exported: SHP (file containing the polygon vertices themselves) SHX (file containing the index of SHP file) DBF (database file containing values and names of entries in SHP file) PRJ (projection file containing information about how to interpret the polygon vertices' coordinates). Neither file has any way to propagate drawing attributes to GIS software (line thickness/color, fill color, transparency, etc.) - it falls to the GIS software to either automatically assign the graphical attributes to the shapefile entries by value, or offer the user a way to assign them. 97

98 Batch Processing SlantView allows users to Process many Datasets automatically, and output KMZs, GeoTIFFs, shapefiles, and/or Workspace files using a batch script. Use the format: SlantView.exe -i"(path to data folder)" -p2 -w -k -g -s -o"(path to desired output folder)" -X The options for Processing are: -i"path" or -I"path" : path to source image folder -o"path" or -O"path" : path to the folder for output products -k or -K : save maps in the KMZ file -g or -G : save maps in GeoTIFF format -s or -S : save maps as shapefiles -w or -W : save the workspace -p0 or -P0 : Weed Detection processing mode -p1 or -P1 : Population processing mode -p2 or -P2 : Stress Analysis processing mode -x : clean up after Processing but do not exit -X : clean up after Processing and exit -e4 or E4 : export images formatted for Pix4D processing, this is a separate processing type and cannot be combined with other options (sample cmd: SlantView.exe - i"c:\slantrange\data\ t185916" -e4 -o"c:\slantrange\data\exportproducts\" -X The following is an example batch script that Processes four Stress Analysis Datasets, saving Workspaces, KMZs, GeoTIFFs, and shapefiles for each: SlantView.exe -i"c:\slantrange\data\ t185916" -p2 -w -k -g -s - o"c:\slantrange\data\exportproducts\" -X SlantView.exe -i"c:\slantrange\data\ t193128" -p2 -w -k -g -s - o"c:\slantrange\data\exportproducts\" -X SlantView.exe -i"c:\slantrange\data\ t185916" -p2 -w -k -g -s - o"c:\slantrange\data\exportproducts\" -X 98

99 SlantView.exe -i"c:\slantrange\data\ t203810" -p2 -w -k -g -s - o"c:\slantrange\data\exportproducts\" -X 99

100 Update firmware Your SLANTRANGE representative will notify you when firmware updates are released. Instructions on performing the update are slightly different for 2p and 3p systems. For 2i and 2p sensors 1. While you have an active internet connection, open SlantView (make sure you have the latest SlantView version available here) and log in with your username and password from analytics.slantrange.com by clicking Settings-User account in the SlantView toolbar. 2. After successfully logging in, connect the sensor to your computer via ethernet cord and power the sensor. 3. Once the AIS lights begin blinking green, select Sensor-Update firmware from the SlantView toolbar. If SlantView is not able to connect to the sensor, try temporarily disabling your Windows firewall or other security software, or set up an exception for SlantView. For 3p sensors Check out this short video on how to update the firmware on your 3p sensor 1. Download the 3p Toolbox app from analytics.slantrange.com, available here. 2. Open the 3p Toolbox program, and with an active internet connection, select Firmware Update, then enter your username and password from analytics.slantrange.com. 3. Once the latest firmware is downloaded, you will be prompted to connect to your sensor's WiFi. Turn on your 3p sensor. 4. Once you see the solid green, yellow, and red lights on the AIS, open your computer's WiFi settings, and select the network name with sensor serial number, formatted as 0003SL3P####. Enter the WiFi password for the sensor: "slantrange3p". 5. Once the 3p Toolbox connects to the sensor, the firmware update will be applied. After a successful update, power down the sensor. Upon next boot up, the sensor is ready to fly! 100

101 101

102 Manually Adjust Boresight Alignment The Boresight Alignment tool was developed for use with the 1p (4000C) system. SLANTRANGE 2p systems have inertial measurement units that automatically account for changes in aircraft orientation. SlantView automatically aligns the images from all sensors, but in some cases, images in the Map Window may appear misaligned. This could be due to gusting wind, improper sensor mounting, or aircraft movement during bootup calibration. If you notice misalignment of images in a map, try using this tool to improve alignment. Click the Sensor-aircraft boresight alignment button (second from the right on the toolbar, boxed in red), or right-click an image in the Map Window to bring up a menu with the Boresight alignment option, shown in Figure 1. Figure 1: Manually adjusting boresight alignment 1 Selecting this option will prompt the display of two adjacent overlapping images in the Image Window, as shown on the right of Figure 2. To perform manual alignment, the user should select an image from the Map Window that has a feature of interest (such as the pivot seen on the right of Figure 1 and Figure 2). It may take a few tries to find two images that share the feature. Once two images with the same feature are selected and displayed in the Image Window, press the F1-F4 keys to make individual adjustments to roll, pitch, heading bias, or heading variation settings. In general, adjusting roll slides the images side to side, adjusting pitch slides the two images up and down, and adjusting heading bias and variation rotate the images relative to one another. Once a particular adjustment is selected for editing, use the scroll wheel of your mouse to increase or decrease the offset. If the user is without a mouse or using a touch screen computer, the buttons in the top right can be used to adjust roll, pitch, heading bias, and heading variation. The outward facing arrows on top of the menu apply negative heading variation and the center facing arrows directly below apply positive 102

103 heading variation. The clockwise arrow applies positive heading bias and the counter-clockwise arrow applies negative heading bias. In the directional pad at the bottom of the menu, the up and down arrows adjust pitch, and the left and right arrows adjust roll. Applying the four types of adjustments will allow you to precisely overlap the feature of interest in the two images, as shown on the right of Figure 2. Once you are satisfied with the position of the overlapping images, click the Settings Processing tab or the cog shaped button in the toolbar, and you will see your new degree offsets populating the boxes within the Boresight offsets, degrees: section of the Settings window shown on the left of Figure 2. Clicking Recalculate maps (circled in red) will apply the boresight offsets (boxed in red) to the data. Figure 2: Manually adjusting boresight alignment 2 103

104 Native Resolution Mosaic Stitching If native resolution fully stitched mosaics are needed for your workflow, all SlantView licenses include the free Lite license for exporting your data for processing in Pix4D, Drone Deploy, and other native resolution mosaic stitching tools. Keep in mind, the industry standard to generate native resolution stitched mosaics of highly aliasing-prone agricultural image data is minimum 80% overlap, resulting in 4x the flight time and 16x the raw data. Navigate to Data--Export--Options to format the image exports as needed. The required formats for importing data to Pix4D and Drone Deploy can be selected with the checkboxes. General options Convert pixel values to reflectance: When selected, image exports are reflectance values rather than intensities. For green crops, the sensor's red and green bands often appear very dark due to the reflectivities typically below 15%. Trim images to valid content: When selected, only the image content shared between all four bands of the SLANTRANGE sensor will be included in the exports. When unselected, black borders are visible around each image, showing the border pixels visible in 3 or fewer of the 4 images planes. Use JPEG compression: When selected, image exports will be JPEG compressed with a quality factor of 90. Targeted Formats Pix4D For your SLANTRANGE images to import correctly into Pix4D, the Convert pixel values to reflectance option must selected, and the Trim images to valid content option must be unselected. This settings profile is selected automatically once the Pix4D box is checked. If these options are modified, the Pix4D box will automatically uncheck. Drone Deploy 104

105 For your SLANTRANGE images to import correctly into Drone Deploy, both the Convert pixel values to reflectance and Trim images to valid content options must be selected. As with the Pix4D Targeted exports, the box will uncheck automatically if this settings profile is modified. Currently, a Drone Deploy Pro subscription is needed to upload and stitch SLANTRANGE Imagery via the link at the bottom of the upload images tab: 105

106 Combine Multiple Flights If you are flying several hundred acres and require battery changes to cover the field, rather than turning off the SLANTRANGE system and having to combine multiple Datasets manually, leave the SLANTRANGE system on throughout the return to base, battery change, and return to the field. SlantView is able to recognize the pit stop, and data quality will not be affected. This is possible if you are using a dual battery setup on your aircraft, or have a separate battery powering your sensor. If the sensor is being powered by onboard aircraft batteries rather than its own external battery (and the aircraft is using at least 2 onboard batteries), replace the aircraft batteries one at a time. As long as one battery is connected, the sensor will stay powered on. If a single onboard battery is powering your sensor and your aircraft, you will have to turn off the sensor, remove the battery, connect the new battery, boot the sensor, and then continue the mission, producing two seperate Datasets. To combine two Datasets, copy the images into one folder and combine the gpsindex.log files. The images will sort themselves into chronological order automatically based on their given timestamps. The gspindex.log files must be combined in chronological order manually; just copy and paste the text from a gpsindex file into the bottom of the gpsindex file from the prior flight using a text editor such as Notepad on Windows. Remember, for SlantView to open a Dataset, the folder must be named with the 15 character date/time identifier. For a combination Dataset, that does not have a unique name assigned by the sensor, just copy the folder name of one of the component Datasets and change the last digit. The grey.tif and sensor_settings.json files from multiple Datasets will overwrite each other when copied into the same folder, just make sure at least one copy of each of these files (from any of the component Datasets) is included. 106

107 Calibrated vs. Uncalibrated Imagery Ambient light conditions change constantly throughout the day and growing season due to weather, angle of the sun, etc. Uncalibrated sensors are useful for taking snapshots in time of the conditions of a given field, but accurately comparing reflectance data from hour-to-hour, month-to-month, and season-toseason is impossible unless the changes in ambient illumniation have been quantified and accounted for. The patented calibration techniques used by the SLANTRANGE system measure incident light via the AIS, and calibrate the data so that stress conditions can be accurately compared across datasets gathered in all lighting conditions. The SLANTRANGE Stress measurement is an absolute measurement that is a ratio of the four spectral bands measured by the sensor. It is our proprietary formula for identifying stress conditions with higher sensitivity and accuracy than can be achieved from industry standard two-band NDVI measurements. Given the variety of crops that our customers measure with their sensors, the system has to be adjustable to display a broad range of reflectivity profiles and stress conditions in an intuitive map. For example, say the sensor measures the reflectivity profile of healthy mature corn and computes stress values between.1 and.2. In another corn field, at an earlier growth stage, in a different part of the world with different soil, the sensor measures stress values between.4 and.5. In these extreme examples, the regions of corn with stress values of.1 and.4 are the healthiest areas from their respective fields. A particular field's conditions must be considered when comparing stress between different crops in different growth stages. At the moment, there is no library of multispectral reflectance signatures for stress conditions in every variant of every crop, in every type of soil, across the world, to reference when assessing the level of stress and causes of stress, in your field. The Stress map shown in SlantView shows the range of Stress values across a field and makes it easy to isolate areas that may require immediate attention or action, but determining what the Stress values measured by the sensor actually correspond to (such as a specific pest, nutrient deficiency, disease etc.) is a question best answered by the agronomist or grower. The SLANTRANGE system can be referred to as a "rapid scouting tool", and the agronomist and grower are still an integral part of interpreting the data. We allow users to adjust the application of our Stress Color Scale to the actual Stress values measured by the sensor (this can cause confusion about the absolute nature of the stress measurement) to show the differences in the measured Stress values in an intuitive way. Typically, we adjust the color scale to show the least stressed areas in shades of green and the most stressed areas in shades of red. To easily compare one field to another, the user simply has to make sure that a particular shade of green corresponds to the same measured stress value in both SlantView Map Windows using the buttons next to the color scale to make incremental adjustments (e.g. setting the darkest shade of green to a stress value of.1). 107

108 The maps in the center, and right of Figure 1 show non-intuitively adjusted color scales. In center of Figure 1, the majority of the stress data (with measured stress values of.1 to.35) is mapped across shades of green. The areas of the field with the highest stress are shown in light green, and are difficult to distinguish from the healthy areas, also shown in green. Similarly, the example on the right has a color scale with the stressed and healthy areas mapped across shades of red. The well adjusted color scale of the map on the left intuitively shows the range of stress values measured in the field, with stressed areas clearly in yellow and red, and the healthy areas in green. Figure 1: Stress color scale adjustments 108

109 Measure NDVIs and Reflectances of custom areas A custom region drawn in the Image Window may be used to measure the reflectance and NDVI of specific areas within a high resolution image. Transfer control to the Image Window, hold down the control key, then left-click and drag to draw a region. If you are looking for a particular reflectance signature to measure, use the number keys 5-9 on the keyboard while active in the Image Window to view images in color (key 5), or in the 550nm channel (key 6), 650nm (key 7), 710nm (key 8), or 850nm (key 9). Figures 1 through 5 below show the same custom region and its reflectance statistics, in the false color view, and each of the four bands. Figure 1: Reflectance and NDVI measurements for a custom region, color image displayed Figure 2: Reflectance and NDVI measurements for a custom region, 550nm image displayed Figure 3: Reflectance and NDVI measurements for a custom region, 650nm image displayed 109

110 Figure 4: Reflectance and NDVI measurements for a custom region, 710nm image displayed Figure 5: Reflectance and NDVI measurements for a custom region, 850nm image displayed 110

111 Use SlantView to Scout a Field This is an example using SlantView with Google Earth to scout and ground truth an orchard. By creating Features and SlantView map overlays in Google Earth, you can walk the field with a phone or tablet in hand to get a closer look at problem areas identified in SlantView. 1. Create Features for the areas you want to scout by right clicking in the Map Window--Feature-- Create: 2. Name the Features: 111

112 3. Once finished defining Features, hover the mouse over one of them and right click. Select Feature-- Export all to--kml. Export the maps as well. 112

113 4. Open the KML in Google Earth to check the points. Export a KMZ of individual or all map layers as needed. Both the KML-features and KMZ-maps will be created in a folder called KML within the main dataset folder. 5. Load into Google Earth to see the maps and features 113

114 6. To get the KML and KMZ file onto a handheld device while walking the field, attachments to yourself or share via Google Drive/Dropbox/iCloud etc. 7. To transfer the files offline: For Apple Devices: File transfer can be done through Itunes ( For Android Devices: Connect the device to your PC via USB-microUSB cable and copy the file through the Windows File Explorer: Once you have the files on the Ipad, open the "Files" app and send both files to Google Earth by pressing and holding the file, clicking "share" from the menu that appears above the file, then clicking "copy to Google Earth" 114

115 1. With the Stress Map, Feature annotations, and an active location marker, you can quickly navigate to address the stressed areas 115

116 Generate Variable Rate Prescriptions The following is an example of loading a SlantView shapefile into SMS. The process for loading shapefiles into other Ag and farm management softwares is similar. The dataset here is a weed detection map of an unplanted field, where the grower generated a variable rate prescription to spray weeds prior to planting. 1. Click "Read File(s)" from the main menu, then "Import a file from a generic source", and locate the file. 2. The most confusing part is identifying and defining the metadata. At first, a blank field boundary is all that will show up. 116

117 3. Select "Generic" as the import type, and "Assign Column." 4. Select "Use as Attribute" and "Add new." 5. Enter a name, abbreviation, decimal number, and leave units dimensionless. 117

118 6. The name you assign will be a new option in the attribute imports, click OK. 7. Select Next, and finish the rest of the import Wizard (options to clip to field boundaries, enter the appropriate grower/field/year names, etc.) The.shp file will show up in the project hierarchy, click "Create New Map" button above the field outline. 118

119 8. Select weed detections (or whatever the data layer was named) from the Attributes drop down bar in the menu to the right of the map. 9. The color scale applied to the data can be adjusted using the "Edit Legend" button. 119

120 10. Within the Edit legend menu, set the color scale to red-yellow-green and uncheck the "visible" option for the green boxes where there are no weeds (value 0.0). 120