Image processing by self-generated spatial and spectral masks

Parallel, analog image processing techniques are described which enable images to be decomposed into subimages that contain objects that are spatially and spectrally equivalent. The decomposition is achieved by generating binary aperture masks directly from the image, with the aid of predetermined signature functions. The versatility of this approach is experimentally demonstrated for some simple scenes. The technique holds promise for on-board, computer-controlled processing of remotely obtained images.