Jose Gomez-Dans wrote:
I have a raster file and a vector file. I want to assign to each pixel
in my raster size a value derived from the vector file (which are
polygons). To do this, I want to test that the centroid of my pixel lies
within a given polygon. In the past, I have done this gdal_rasterize-ing
the polygon file, and going through numpy arrays. It is a bit
cumbersome, and the rasterization is not really needed with OGR.
Perhaps cumbersome, but in fact, at some level, you are indeed doing a
rasterization with the P-in-P testing, as you are working with ALL the
pixels in your raster. I bet it would be a lot faster to rasterize the
polygons once, rather than doing point-in-polygon tests for every single
raster point.
Think of it this way:
With rasterize, you are first finding ALL the points in a given polygon.
With P-in-P testing, you are finding if one point is in a given polygon.
If you had a few points to do, it would be a waste to find them all, but
since you are, in fact, doing all the raster points, why not only do it
all at once?
-Chris
--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
[EMAIL PROTECTED]
_______________________________________________
gdal-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/gdal-dev
