6

My question is similar to Extracting raster values at points using Open Source GIS?, but perhaps more complex.

How to extract the attributes of clustered cells around a point/points?

For example, I have many vegetation plots in 100m x 100m and have to examine the spectra at the points over SPOT Data (satellite image) which has 10 m resolution.

  • So you need to extract satellite image data of 100x100 m areas? – underdark Feb 24 '11 at 12:54
  • 1
    Can you attach an image to illustrate the situation? – Pablo Feb 24 '11 at 13:21
  • @Wim Are you familiar with the zonal operations in ArcGIS? webhelp.esri.com/arcgisdesktop/9.3/… – whuber Feb 24 '11 at 15:22
  • @Wim and @Bill, for version 10, zonal operations start here help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//… – user681 Feb 24 '11 at 16:23
  • thanks for commenting. please let me explain more..the plots in my example are represented in point feature, so there are many points. And I have another image to which I'd like to extract the certain cells values (3x3 or 5x5 cells around each point). These values would be inserted to attribute table of points (plots). @whuber: thanks for the idea. but i'd like to have the cell values, not only the statistics. – wnursal Feb 24 '11 at 16:47
2

One way to proceed using SAGA GIS, but I'm sure other programs may do this also: open your points and grid:

  1. convert you points to gridcells (module shape to grid), in the same grid system as your other grid

  2. buffer the grid you just created (module: grid buffer)

  3. use the module grid values to points and select both grids (buffer and original one), you will now get a point shapefile which contains in one column the id of the original point (which you buffered) and in another one the values inside the buffer.

http://saga-gis.org/en/index.html

  • The same approach can be used in other GIS software: create vectors which overlap your area of interest, then intersect these vectors with your rasters of interest and inspect the values within. – scw Feb 27 '11 at 5:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.