Image Analysis and Neural Networks

Abstract

In the applications presented so far, there was a strong requirement for imaging algorithms that implement basic functions in human visual perception like edge detection or shape from stereo. The idea was to employ these techniques as sophisticated solutions for complex modeling problems. However, aside from these strictly visual analysis techniques, there is another subject that is even more general in the sense that it stands for intelligent behavior, such as decision making and classification. Each of us knows that human beings perform these tasks very well even if the underlying facts are unclear or fuzzy. In visualization, for instance, segmentation and cluster analysis are extremely important for the prefiltering of large data sets. Moreover, in imaging and even more in artificial intelligence in general [24] one of the mainstream working areas turns out to be segmentation and classification.