基于Hopfield神经网络模型的启发式学习算法及其在数字模式处理中的应用

A HEURISTIC LEARNING ALGORITHM BASED ON HOPFIELD NEURAL NET WORKS MODEL AND USED FOR PROCESSING DIGITAL PATTERN

摘要: 本文给出了一个基于Hopfield神经网络的启发式学习算法．使用这个算法对6个分别被0．1和0．5随机噪声干扰的数字模式成功地进行了校正处理，得到了较好的结果．

Abstract: A heuristic learning algorithm based on Hopfield neural net is given in this paper. Using this algorithm, wecorrect successfully six digital patterns corrupted respectively by random noise 0.1 and 0.25 of each pixel.