Module 3 Introduction to GIS. Lecture 8 GIS data acquisition

Transcription

1 Module 3 Introduction to GIS Lecture 8 GIS data acquisition

2 GIS workflow Data acquisition (geospatial data input) GPS Remote sensing (satellites, UAV s) LiDAR Digitized maps Attribute Data Management Data verification Database management Exploratory Analysis Attribute and spatial data queries Geovisualization Data Analysis Vector and raster data analysis Terrain mapping Spatial interpolation Network analysis Geovisualization (maps) Chang, 2014, p.8-9

3 Unmanned aerial vehicles GIS data acquisition GPS Remote sensing Aerial photography Digitised maps LiDAR

4 GIS data acquisition - vector GPS Digitised maps

5 GIS data acquisition - raster Aerial photography Remote sensing (Satellite images)

6 Remote sensing

7 Remote sensing definitions REMOTE sensor located at a platform, at some distance from the object or area (no physical contact) Platforms: satellites, airplanes, unmanned aerial vehicles drones Sensors: placed in cameras with different characteristics Why this matters for spatial data acquisition? Fieldwork is costly and some places are not easily accessible The person and equipment collecting the data might interfere with the objet or area being studied Acquisition over larger areas, for several time periods

8 Remote sensing definitions SENSING sensor capturing light reflected from the object or area (surface) Why this matters for spatial data acquisition? Surfaces reflect light differently according to their characteristics that allows to differentiate them Energy source (sun) required to illuminate a target. Passive sensor used the sun as source Active sensor own source of energy When receiving light: surfaces reflect, absorb and transmit that light - electromagnetic (EM) energy (or radiation). The EM energy reflected by the surface is detected and recorded by the sensor and converted to digital images (each pixel in the image contains specific data related to the way the surface reflected light).

9 Remote sensing definitions EM spectrum Visible light: 0.4 µm (blue) to 0.7 µm (red) Wavelengths measured in microns or micrometres (µm) 10-6m.

10 Remote sensing definitions EM spectrum / Specific data Surfaces are EM spectrum selective Sensors are EM spectrum selective Digital/satellite images spatial data SENSOR SURFACE Each surface has a spectral signature (Allows differentiate trees, buildings, water ) Each sensor has different bands Specific data: each pixel records the amount of energy emitted for each band, converted afterwards into thematic data

11 ttp:// Remote sensing definitions Spectral signatures

12 Remote sensing images Digital images spatial data acquisition using remote sensing systems

13 Remote sensing images Features to consider for remote sensing images: SPATIAL RESOLUTION (area extent and pixel size) (platform & sensor) SPECTRAL RESOLUTION (number of bands) (sensor) TEMPORAL RESOLUTION (acquisition frequency over the same location) (platform & sensor) Choosing the right image depends on the application but it is often a trade-off of these resolution features and the budget available.

14 Remote sensing images Spatial Resolution Spatial resolution: pixel size Usually expressed in meters (for example, a 30 m resolution means that two objects, thirty meters long or wide, sitting side by side, can be resolved on the satellite image) High resolution: pixel size from 0.4 to 4 m Medium resolution Low resolution: pixel size from 30 m to more than 1 km

15 Remote sensing images Spatial Resolution 30 m 10 m 2 m

16 Remote sensing images Spatial Resolution 1 m 10 m 30 m

17 Remote sensing images Spectral resolution Spectral resolution: number of spectral bands + regions of the EM spectrum In terms of the spectral regions used, the remote sensing systems can be classified into: Optical (include visible, near infrared, and shortwave infrared systems) Thermal Synthetic aperture radar (SAR) In terms of the number of bands recorded, optical/thermal systems can be classified into: Monospectral or panchromatic (single wavelength band) Multispectral (tens of spectral bands) Hyperspectral (hundreds of spectral bands)

18 Remote sensing images Spectral resolution /Panchromatic 1m IKONOS panchromatic image of Bogota, Colombia Image credit LAND INFO Worldwide Mapping, LLC, includes material Copyright DigitalGlobe - Longmont, Colorado 10 m SPOT HRV panchromatic image of San Francisco, USA (

19 Remote sensing images Spectral resolution /Multispectral 4 m IKONOS multispectral image of Cape Town, South AfricaI Planetary Visions Limited 1.5 m SPOT 6/7 multispectral image of the Kiribati Republic Copyright: Airbus DS 2015

20 Remote sensing images Spectral resolution /Hyperspectral 5m HyMap hyperspectral imagery from the Willouran ranges in South Australia. Carbonate materials are shown in blues and purples while siliclastic materials appear as oranges and reds. Vegetation is displayed as green (J.L. Keeling and A.J. Mauger, 2000: "Application of Airborne Hyperspectral (HYMAP) Data to Map Variation in Carbonate Facies in Proterozoic Skillogalee Dolomite, Willouran Ranges, South Australia." 10th Australasian Remote Sensing and Photogrammetry Conference, Adelaide, 2000; )

21 Remote sensing images Temporal resolution Revisiting frequency of a satellite sensor for a specific location High temporal resolution: scans the same location less than 24 hours to 3 days after the first scan. Medium temporal resolution refers to 4 to 16 days. Low temporal resolution refers to 16 days.

22 Examples of satellite images The WorldView-3 Satellite Sensor provides 31 cm panchromatic resolution ( nm), 1.24 m multispectral resolution ( nm) and 3.7 m short wave infrared resolution ( nm). It has an average revisit time of less than 1 day.

23 Remote sensing images Visualization how do we choose to display spectral data for visual analysis

24 Image visualization Image visualization: assigning the right "colours" to the image bands (colour composite image) to improve image interpretation Colour composite image: associating each band (not necessarily the visible band) to a separate primary colour (red, green and blue - RGB) TRUE COLOUR FALSE COLOUR

25 Image visualization False colour composite Common false colour band combination for image interpretation (identify healthy vegetation): Near infrared, green, and red bands displayed as red, blue and green Vegetation appears in different shades of red depending on the types and conditions of the vegetation, since it has a high reflectance in the NIR band. TRUE COLOUR FALSE COLOUR

26 Remote sensing visualization Remote sensing software (image viewing): red, green and blue colours assigned to any of the bands collected by the sensor. Satellite-derived remotely sensed images are often viewed as false colour composites: red band to show near-infrared, green band to show visible red and blue band to show visible green. False colour composite image (Landsat Thematic Mapper image of Morro Bay, California)

27 LiDAR

28 LiDAR - Light Detection and Ranging (3D) laser scanning method used to map the surface of the Earth. Often mounted in an aircraft or any other flying device (although it is also possible to have ground mounted devices), the LiDAR device uses light in the form of a pulsed laser to measure ranges (variable distances) to the Earth. The light pulses combined with GPS data (altitude, latitude and longitude) recorded by LiDAR system generate precise, 3D data for the shape of the Earth and its surface characteristics. The data produced is in a 'point cloud' format: 3D array of points, each having x, y and z positions relative to a chosen coordinate system. A single LiDAR survey can easily generate billions of points totalling several terabytes.

29 LiDAR - Light Detection and Ranging

30 Cool LiDAR applications LiDAR helps revealing hidden cities LiDAR helps creating exciting music videos LiDAR helps natural park management & cultural heritage

31 Unmanned aerial vehicles (UAVs) Remotely controlled aircraft systems able to fly at low altitudes, carrying different sensors from a camera to a LiDAR If the image resolution is not high enough to see exact areas of devastation or change, coverage of an entire affected area is not available, or imagery is simply too expensive to acquire, then an analysis will be difficult to complete. The generally low-cost high resolution image capture capability of UAV s creates the potential for them to fill the data gap between satellites, ortophotos and ground surveying. By attaching different sensors, it is possible to gather a wide range of environmental data (air pollution, land cover surveying, wildlife recording, heat mapping). They can also assist with emergency management operations, eliminating the humanrisk factor from the operation while also sending back real-time information.

32 UAV s in action Emergency management Conservation management:

33 More about GIS analysis next week SCI103 notes: Start planning Assessment 5 (any questions yet?)