0

The green values have a COUNT attribute of 13418

The green values have a COUNT attribute of 13418. I will want to use a tool that lumps the COUNT attribute and reclassifies based on values. But first it appears I will need to break up the raster cells into individual values using a tool? I am learning raster, in the past I only used vector.

Improve this question
5
  • Your question has many different possible interpretations. Could you please tell us (a) whether you are making a distinction between the Count attribute and a "value" and (b) what it means to "break up" raster cells?
    – whuber
    Commented Oct 9, 2013 at 16:10
  • I need a tool that breaks up the Forestland into individual values based upon contiguity. Then use a tool that looks at the “COUNT” attribute (total cells) and reclassifies based on this value. Eventually I want to be in a Contiguous Forest Block (the green cells) that is comprised of at least 1000 Unbroken Cells.
    – MichaelGEO
    Commented Oct 9, 2013 at 16:22
  • 1
    I'm still not sure what you mean by "break up" the forestland. Are you perhaps describing a RegionGroup operation? BTW, please make explicit the software you would like to use (from the screen shot I would guess it is Spatial Analyst).
    – whuber
    Commented Oct 9, 2013 at 16:48
  • The counts of the raster data are concentrated as one value (the green box with a one). I need the values dispersed and individualized based on contiguity (where they touch another).
    – MichaelGEO
    Commented Oct 9, 2013 at 17:23
  • Sorry for any unclearness, and yes only spatial analyst tools
    – MichaelGEO
    Commented Oct 9, 2013 at 17:36

1 Answer 1

Reset to default
1

I believe you are looking for the Region Group tool in the spatial analyst toolset. It allows you to "regroup" your clusters of like values, while allowing you to ignore NoData cells. You can also specify the type of neighborhood. The raster attribute table will then contain a count of cells in each group. I hope it helps.

Improve this answer
1
  • Thank you! You all were correct in suggesting the RegionGroup operation, it broke up my resulting grid into individual values; that I later reclassified and factored out cells with less than a '1000' cell count.
    – MichaelGEO
    Commented Oct 9, 2013 at 18:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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