Implementation of artificial neural networks for recognition of target and clutter images

It is presented the development and comparison of two systems for the recognition of the targets and of clutter by means use artificial neural networks (ANN). Images SAR (synthetic aperture radar) of targets and clutter were obtained using a detect system of potential targets CFAR (Bhanu, 2003). Three forms of ANN training were used. The first one consists of the net training with images that are not contaminated by noise. The second one is based on training of the net with images that have been contaminated by noise. And, the third form used training with different level of noise. The results show the capacity and differences of the systems in recognition of radar images.