基于灰度共生矩阵的模糊神经网络FuCMAC在图像分割中的应用  

Head MRIs Segmentation Based on a Novel Fuzzy Neural Network FuCMAC and Graycomatrix

作　　者：程起才[1] 王洪元[1] 

机构地区：[1]江苏工业学院计算机系,江苏常州213164

出　　处：《江南大学学报（自然科学版）》2007年第6期842-845,共4页Joural of Jiangnan University (Natural Science Edition)　

基　　金：江苏省高校自然科学基础研究基金项目(06KJD520048)

摘　　要：文中提出了一种改进的CMAC神经网络(FuCMAC),与经典的FCMAC相比,其逼近精度得到提高,解决了CMAC系列神经网络逼近精度不高的弱点.在颅脑磁共振图像分割仿真实验中,把当前像素点的子图像的纹理特征和该像素点的灰度值作为该像素的特征向量,将该特征向量作为神经网络的输入,结果表明其具有较高的分割准确性.A novel fuzzy CMAC neural network, called FuCMAC, is presented in this paper. Compared with the classical FCMAC, its approximation accuracy is enhanced and the common fault of the series of CMAC neural network can be solved. In the simulation for the segmentation of the head magnetic resonance images, the gray value of the current pixel and texture features of neighbourhood of the current pixel are used as the inputs for FuCMAC. The simulation also demonstrates that its effect of segmentation is better than the classical FCMAC.

关 键 词：灰度共生矩阵 磁共振图像 模糊小脑神经网络 图像分割 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]