Image recovery and segmentation using competitive learning in a neighborhood system

In the previous work based on the principle of low-level mammalian visual system that deals with image restoration and segmentation problems from a more direct and easily understandable and acceptable aspect, we (1996) developed a new algorithm. Incorporating the competitive learning method, this algorithm yields improved performance over previous studies in synthetic image restoration. Within the framework of Markov random fields (MRF), our assumption is that the observations lie in an MRF. The image recovery problem is transformed to the problem of minimization of an energy function. A local update rule for each pixel point is then developed in a stepwise fashion and is shown to be a gradient descent rule for an associated global energy function. This paper deals further with the development and application of this algorithm, and focuses on a comparison of different parameters and noisy images