Finding intersection between two color maps and displaying it

Hi, this is my first time using the image processing toolbox.

Currently I am trying to image process these two maps: http://hurricane.ncdc.noaa.gov/climaps/temp1007.pdf http://hurricane.ncdc.noaa.gov/climaps/rh2307.pdf

I was wondering if there is a way to find all spots on a map of these color coded maps of the US, where the number of days the temperature was above 90 was at least 10 days, and the relative humidity was below 56 percent. The program would then display these spots on the map.

Conceptually I think it would really just be making a series of if statements covering the range of colors that could go together on the map. For example, if a certain spot on the map was green on the humidity map(46-55%) and yellow on the temperature map (15.5-20.4 days above 90) it would be displayed. So in boolean it would translate to: if(humidspot==green && tempspot==yellow), then display spot.

Is this possible to do with the image processing toolbox? I know, pretty long question, but I'd appreciate the help.

Thanks,

Steven

Answer 1

[Walter Roberson]

Well, the image processing toolbox would be useful for finding the Region of Interest, and for doing some image registration.

I would suggest that if you were to use rgb2ind() with the number of defined categories plus 2 (one for the black boundaries, one for the white area outside the map), and then do a simple check to find out which ind corresponds to which category (by comparing the colors in the output color map), then all you would have left is the logical "and" of pair of ismember() tests (comparing the pixel indices to the indices matched to be ones of interest for your purpose.)

Getting the maps to register with each other properly would likely be the most difficult part, provided the color scheme stays constant.