Im trying to analyze and extract patterns from several raster layers, something like what is done for non-raster features. Im interested in learning techniques that describe the distribution of cells across my study region. I see similar aims are achieved using moving window analyses, local point or kernal density or hotspot analysis in arcmap. Problem is, you can’t use any of these arcmap tools for rasters, it generally requires point data. I can’t convert these raster cells to points because the raster layer contains various land cover categories or more than one value. Is there a way to extract the different values from a raster layer to points in a feature layer but retain the categories that exist in the raster? Or, is there any tool I can use to analyze or determine patters in rasters. Also open to using R to do this? Thanks in advance!
dear lavinia,
for categorized features u should firstly reclassify the features to define attribute table, and then u can convert the categories to the point distinctly.
Perhaps you can try to process the image with Python. There are many resources online. For example, you may want to take a look at sci-kit-image (https://scikit-image.org/). If you have a lot of data, e.g., a temporal series, maybe you would like to try Google Earth Engine (https://earthengine.google.com/). Success
Dear Lavinia,
I suggest to consider the raster package (https://cran.r-project.org/web/packages/raster/vignettes/Raster.pdf) in R. Alternatively, use QGIS. Extracting the values of a raster at specific points is achieved by a single commandin R:
raster::extract(your_raster, your_xy).
There are also packages/functions for computing kernel density maps, to assess autocorrelation etc etc. If any function you need is missing (unlikely), you can easily write it yourself. I suggest to learn R by runnimg it in Rstudio.
ROSA,
es extraño que un raster te maneje mas de un solo atributo. en tod caso puedes probar codificando los atributos tal como lo hacia arc-info.y luego recoviertes los codigos en atributos
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning