Segmentation of FLIR images by Hopfield neural network with edge constraint

A segmentation algorithm of forward-looking infrared (FLTR) images by Hopfi eld neural network (HNN) with edge constraint is presented. An evaluation c riterion based on distinct edge pixels is used to examine the segmentation results by HNN under different initial assignment of probabilities. Thus, t he good segmentation result can be achieved by automatically adapting initi al assignment of probabilities to reach the optimal or suboptimal solution of the evaluation criterion. To determine appropriate weights of the object ive function and the constraint condition in the energy of HNN, a criterion with respect to the constraint condition is proposed. Experimental results with real FLIR images are given. (C) 2001 Pattern Recognition Society. Pub lished by Elsevier Science Ltd. All rights reserved.