2

I have a raster containing different pixel values and I am interested in having squares drawn around areas that are at least X m^2 and containing at least Y type of pixels.

The data looks like this, a coverage with the following attributes:

OID     Value     Count
0       1         123123
1       2         3342
2       4         2342342
3       5         23423423
4       7         2342323

1) I would like to identify (and draw) squares of a certain area around places where there exists at least a certain amount of a pixel value.

2) Squares may overlap.

It would be nice to have the resulting squares output to a separate layer with each square being a unique feature with the following attribute data (number / % of pixels of each value, area, dimensions...).

Improve this question
2
  • 1
    Not enough info. Can squares overlap? How many per group?
    – FelixIP
    Jul 14, 2015 at 5:54
  • question edited.
    – geokrowding
    Jul 14, 2015 at 6:36

1 Answer 1

Reset to default
1

No script solution

  1. Convert raster to binary, assigning 1 to value of interest, 0 for the rest
  2. Focal statistics, SUM, with appropriate size square.
  3. Raster with value => y to points
  4. Buffer of points
  5. Polygons from individual buffer extents
Improve this answer
2
  • That will generate the # of pixels per square, for sure, but will generate a square of a chosen size for every point in the coverage. I am only looking to generate squares that satisfy the condition of at least X pixels of a certain value, and at least Y area (drawing a square to suit the largest extent that satisfies the minimum % of pixel values of interest). PErhaps I shall amend the question to better clarify that. Thanks for your attempt and it provides at last part of solution :)
    – geokrowding
    Jul 14, 2015 at 17:02
  • what about the minimum bounding geometry tool (under data management). Anyone have any experience with that or know if it could be rolled into this analysis?
    – geokrowding
    Jul 14, 2015 at 17:30

Your Answer

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

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