Dirty REMOTE SENSING Week 2 Interpreation

Transcription

1 Dirty REMOTE SENSING Week 2 Interpreation Earthobservation.wordpress.com Stuart Green Stuart.Green@teagasc.ie

2

3 AERIAL PHOTOGRAPHIC INTERPRETATION

4 Images of the earth are map like representations of the earth s surface based on reflections of electromagnetic energy from vegetation, soil, water, rocks, and structures. Yet such information is not presented to us directlythe information we seek is encoded in the various tones and textures we see on the image. To translate images into information we must apply a specialised knowledge-knowledge that forms the field of image interpretation, which we can apply to derive useful information from the raw, un-interpreted images we receive from remote sensing systems. These definations are taken from Campbell 2002: ontcover&source=gbs_navlinks_s#v=onepage&q=&f=false

5 What are we looking for? Knowledge of the subject :Accurate interpretation requires familiarity with the subject of the interpretation. Image interpreters should be equipped with a broad range of knowledge and experince across earth and biological sciences.

6 The Human Image Processing Continuum natural perception (looking) -> abduction (hunch) -> convention (experience/teaching)-> cognitive (Understanding) tml

7 First Questions Where am I Looking? What system took the image? When was it taken?

8 Remember. (1) Remotely sensed imagery usually presents an overhead view. (2) Many remote sensing images use radiation outside the visible portion of the spectrum. (3) Remote sensing images often portray the earth s surface at unfamiliar scales and resolutions.

9 Elements Tone/Colour Size, Shape & Texture Pattern, Shaddow Location, Context

10 1. Image Tone

11 2. Image Texture two definitions> Image texture refers to the apparent roughness or smoothness of an image in a region. Usually texture is caused by the pattern of highlighted and shadowed areas as an irregular surface is illuminated from an oblique angle. Image texture depends on the surface itself and the angle of illumination, so it can vary as lighting varies. Good rendition of texture depends on favourable image contrast, so images of poor or marginal quality may lack the distinct textural differences so valuable to the interpreter. Alt:Texture is the frequency of change and arrangement of tones. This is a micro image characteristic. The visual impression of smoothness or roughness of an area can often be a valuable clue in image interpretation. Still water bodies are typically fine textured, grass medium, brush rough. There are always exceptions though and scale can and does play a role; grass could be smooth, brush medium and forest rough on higher altitude aerial photograph of the same area.

12 Image Texture

13 3. Shadow Shadow is crucial in telling us about relative size and position

14 Shadow

15 Relief inversion We expect images to be illuminated from above and for images to be orienated with north at the topthis can cause problems

16 4. Pattern It must be noted here that pattern is highly scale dependent.

17 Pattern

18 5. Association/context Association is a functional relationship: You can t have a railway station without a railway line

19

20 6.Shape & Size Size can be absolute if you have a scale or it can relative; one thing is bigger than the other

21

22 8. Location Location is about position in landscape, especially in realtion to topography

23 Photomorphic Regions

24 Photomorphic Regions In the first step the interpreter delineates regions of uniform image appearance, using tone, texture, shadow, and the other elements of image interpretation as a means of separating regions. In the second step the interpreter must be able to match photomorphic regions to useful classes of interest to the interpreter.

25 National Grasslands Survey Done by hand digitising of regions. Now can do automatically through Image Segmentation

26 LIST of MAIN satellite imaging systems for Irish use in RS of Land Landsat Terra-aster SPOT EROS IKONOS QuickBird Radarsat ENVISAT ERS worldview IRS DMC

27 IRS (Indian Remote Sensing Satellites) IRS P6 (ResourceSat-1) 17/10/2003» Satellite) ALOS 1 24/01/2006» ALOS (Advanced Land Observing BIRD 1 (Bi-Spectral Infrared Detection) BIRD 1 22/10/2001» Jason (Joint Altimetry Satellite Oceanography Network) JASON 07/12/2001» CBERS (China/Brazil Earth Resources Satellite) CBERS-2 (ZY-1B) KOMPSAT (Korea Multi-Purpose Satellite) KOMPSAT 1 20/12/1999» 21/10/2003» KOMPSAT (Korea Multi-Purpose Satellite) KOMPSAT 2 28/07/2006» CloudSat CLOUDSAT 1 28/04/2006» LANDSAT LANDSAT 7 15/04/1999» Digital Globe Constellation WorldView 1 18/09/2007» LANDSAT LANDSAT 5 01/04/1984» Digital Globe Constellation QuickBird 2 18/10/2001» DMC (Disaster Monitoring Constellation) Beijing-1 27/10/2005» METEOR-3 METEOR-3 24/10/1985» DMC (Disaster Monitoring Constellation) UK-DMCSat-1 27/11/2003» MetOp (Operational Meteorological Satellites) MetOp 1 19/10/2006» DMC (Disaster Monitoring Constellation) NigeriaSat-1 27/11/2003» MFG (Meteosat First Generation) Meteosat 6 20/11/1993» DMC (Disaster Monitoring Constellation) BILSAT-1 27/09/2003» MFG (Meteosat First Generation) Meteosat 5 02/03/1991» DMC (Disaster Monitoring Constellation) AlSAT-1 28/11/2002» MFG (Meteosat First Generation) Meteosat 7 03/09/1997» ENVISAT (Environmental Satellite) ENVISAT 01/03/2002» Monitor-E (Monitor Experimental) MONITOR-E 1 26/08/2005 EO-1» (Earth Observing-1) EO-1 21/11/2000» EOS (Earth Observing System) AQUA 04/05/2002» MSG (Meteosat Second Generation) MSG 1 28/08/2002» EOS (Earth Observing System) GRACE 17/03/2002» MSG (Meteosat Second Generation) MSG 2 22/12/2005» EOS (Earth Observing System) TERRA 18/12/1999» MTI (Multispectral Thermal Imager) MTI 12/03/2000» EROS (Earth Remote Observation System) EROS-B 25/04/2006» OrbView OrbView 2 01/08/1997» EROS (Earth Remote Observation System) EROS-A 05/12/2000» ERS (European Remote-Sensing Satellite) ERS-2 21/04/1995» OrbView OrbView 3 26/06/2003» FORMOSAT FORMOSAT 2 21/05/2001» OrbView OrbView 1 01/04/1995» FORMOSAT FORMOSAT 3 16/04/2006» POES (Polar Orbiting Environmental Satellites) NOAA 16 21/09/2000» GOES (Geostationary Operational Environmental Satellite) GOES 12 POES (Polar Orbiting Environmental Satellites) NOAA 17 24/06/2002» 23/07/2001» POES (Polar Orbiting Environmental Satellites) NOAA 18 20/05/2005» Operational Environmental Satellite) GOES 9 GOES (Geostationary 23/05/1995» POES (Polar Orbiting Environmental Satellites) NOAA 14 30/12/1994» GOES (Geostationary POES (Polar Orbiting Environmental Satellites) NOAA 12 14/05/1991» Operational Environmental Satellite) GOES 11 03/05/2000» POES (Polar Orbiting Environmental Satellites) NOAA 15 13/05/1998» GOES (Geostationary Operational Environmental Satellite) GOES 10 PROBA (Project for On-Board Autonomy) PROBA 1 22/10/200125/04/1997»» RadarSat constellation RadarSat 1 04/11/1995» GOES (Geostationary Operational Environmental Satellite) GOES 13 24/05/2006»» Resurs-DK1 (Resurs - High Resolution 1) RESURS DK1 15/06/2006 Ikonos IKONOS-2 24/09/1999» SciSat (Science Satellite) SCISAT 1 13/08/2003» INSAT (Indian National Satellite) INSAT 2E 02/04/1999» SPOT (Système Pour l'observation de la Terre) SPOT 5 04/05/2002» INSAT (Indian National Satellite) INSAT 2C 06/12/1995» SPOT (Système Pour l'observation de la Terre) SPOT 4 24/03/1998 INSAT (Indian»National Satellite) KALPANA 1 12/09/2002» SPOT (Système Pour l'observation de la Terre) SPOT 2 22/01/1990 INSAT (Indian»National Satellite) INSAT 3A 09/04/2003» IRS (Indian Remote Sensing Satellites) IRS 1C 28/12/1995» TOPSAT TOPSAT 1 27/10/2005» IRS (Indian Remote Sensing Satellites) IRS 1D 29/09/1997» TSINGHUA TSINGHUA 1 28/06/2000 IRS (Indian Remote Sensing Satellites) IRS P4 (OceanSat-1) 26/05/1999» IRS (Indian Remote Sensing Satellites) IRS P5 (CartoSat-1) 05/05/2005»

28 Multiple Bands Remember a satellite doesn t just record an image in RED, GREEN and BLUE light. The satellite is able to record information across a wider spectrum and with more discrimination

29 We will use a Landsat8 OLI image with 11 bands

30 B1- Violet: coastal/aerosol band B2: Blue B3: Green B4: Red B5: Near Infra Red NIR- Plant detector B6&7: Short Wave Infra Red- Soil wetness/rock B8: Panchromatic-Higher Res Black and white B9: Cloud Detector B10&11: Thermal Imaging

31 nd-combinations-for-landsat-8/ Natural Color False Color (urban) Color Infrared (vegetation) Agriculture Atmospheric Penetration Healthy Vegetation Land/Water Natural With Atmospheric Removal Shortwave Infrared Vegetation Analysis 6 5 4

32 Practical Using ARC Map today (remember to start the macs in windows) Use Carlow17.tif and sites file (sites.shp) from blog Open Arc Map, load bookmarks and load Base Images. Visit each of the 6 locations and describe what you can see- try to identify the landcover, landuse and habitat

33 Write out your observation using a table like this Site Number Habitat Type Decsription Alternatively, if you are comfortable with ARCMAP you can edit the sites.shp attribute table For Guidance Use Fosset

34 Load Carlow17.tif visit each of the 6 sites- what is the best band combination to distinguish the site form its neighbours. Here's how to load an image and change bands Or here Try this alternate tutorial

35 3/4 One trick you ll have to learn is how to distinguish between dry vegegation and bare soil

36 4/4 Compare the 6 sites Load the Base Map/Image. What can you see at the six sites in the aerial imagery that you can t see in the satellite? Do you need to change any of your 6 habitat observations? What does shaddow tell you for each of the 6 locations