Using segmentation to automate the interpretation of multispectral images
R. M. Aleev, V. B. Fofanov · Journal of Optical Technology · 2009
This paper discusses the possibilities of segmenting the zones of interest on multispectral images, as applied to two mathematical models of a scene. It is shown that the use of unfounded assumptions when constructing a model can cause false information to be obtained about a scene. Results are given from computer experiments on the segmentation of zones of interest, using trispectral images.