I am working with Landsat scenes for classification on a large area. This area is covered by 10+ images.My work has to be scripted in python for multiple use across different areas in the worldI need to perform an equalization in reflectance in this set of data to have an homogenous result.I am not looking for merging the rasters, 10+ Landsat images would give something unusable to work with.
I assume the best way is to begin with a "master image" whose histogram would be the reference, then I'd like to equalize an adjacent image, by computing a scatterplot over the two histograms, and fit a 2nd degree polynomial. I would adjust the second image according to the first one.
And then, by propagation, I have to get all the pictures on the same "basis" of reflectance than the master image.
But to do that, I need to know what are the existing overlaps in this set, I need to save time, and not trying to perform the algorithm over two images which don't overlap.
I thought about a sort of graph that would link every images having a overlap with each other. The thing is that I am not very aware about how to with graph, and in python, even less.
أكثر...
I assume the best way is to begin with a "master image" whose histogram would be the reference, then I'd like to equalize an adjacent image, by computing a scatterplot over the two histograms, and fit a 2nd degree polynomial. I would adjust the second image according to the first one.
And then, by propagation, I have to get all the pictures on the same "basis" of reflectance than the master image.
But to do that, I need to know what are the existing overlaps in this set, I need to save time, and not trying to perform the algorithm over two images which don't overlap.
I thought about a sort of graph that would link every images having a overlap with each other. The thing is that I am not very aware about how to with graph, and in python, even less.
أكثر...