论文部分内容阅读
在本文中我们提出了将细胞神经网的模板设计与对Ｇｉｂｂｓ统计图像模型的分析结合起来，形成一个用于解决诸如图像恢复、分割、边缘检测、重建等问题的细胞神经网络，新的通用设计方法──能量函数法．这种方法将权的学习、硬件退火等方法有机地结合起来，从而合理地得到细胞神经网络模板形式，较好地解决一些图像处理问题．
In this paper, we propose to combine the template design of cellular neural network with the analysis of Gibbs statistical image model to form a cellular neural network for solving problems such as image restoration, segmentation, edge detection and reconstruction, a new universal design Method ─ ─ Energy Function Method. This method combines the study of right, hardware annealing and other methods organically, so as to get the form of cellular neural network template reasonably and solve some image processing problems well.