基于自组织特征映射网络SOM的图像分割方法  

Image-Segmentation based on SOM neural network

作　　者：摆晔[1] 李志勇[1] 蒋慕蓉[1] 

机构地区：[1]云南大学信息学院,云南昆明650091

出　　处：《云南大学学报（自然科学版）》2007年第S2期275-278,共4页Journal of Yunnan University(Natural Sciences Edition)

基　　金：云南省应用基础研究计划项目(2007F153M)

摘　　要：阈值法是图像分割中常用的有效分割技术之一,但是阈值的选取对分割结果影响很大,特别是对于生物医学图像,分割结果往往不能满足要求.本文将SOM神经网络应用到图像分割中,利用其自组织学习的能力自动获取阈值,其结果不仅优于传统阈值法,而且也优于竞争型神经网络.Threshold method is commonly used in image segmentation as a effective segmentation technique,but the selection of threshold has a great impact on the results,especially for biomedical images,the segmentation results often do not match the requirements.This paper used SOM neural network applications to image segmentation,uses of its self-organizing learning capabilities to Automatic Generate threshold,the result is not only superior to the traditional threshold,but also superior to the competitive neural networks.

关 键 词：SOM神经网络 图像分割 阈值法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]