首页 | 官方网站   微博 | 高级检索  
     

基于PCNN模型的区域生长图像分割
引用本文:刘莉,谈文蓉,傅春常.基于PCNN模型的区域生长图像分割[J].西南民族学院学报(自然科学版),2014(3):434-438.
作者姓名:刘莉  谈文蓉  傅春常
作者单位:西南民族大学计算机科学与技术学院,四川成都610041
基金项目:四川省应用基础研究计划项目《基于PCNN的医学图像处理关键技术研究》(项目编号:2013JY0188)
摘    要:提出了一种结合区域生长算法和脉冲耦合神经网络进行图像分割的方法.该方法将待分割图像的像素点映射为PCNN模型中的神经元,把改进的脉冲耦合神经网络模型的点火频率同区域生长的理论结合起来进行图像分割.实验表明该方法分割的图像与传统的分割法相比具有边缘信息更加完整,区域划分更加准确,分割效果更能符合人眼视觉的识别特征.

关 键 词:数值图像处理  图像分割  区域生长  PCNN模型

Region growing image segmentation based on PCNN model
LIU Li,TAN Wen-rong,FU Chun-chang.Region growing image segmentation based on PCNN model[J].Journal of Southwest Nationalities College(Natural Science Edition),2014(3):434-438.
Authors:LIU Li  TAN Wen-rong  FU Chun-chang
Affiliation:(School of Computer Science & Technology, Southwest University for Nationalities, Chengdu610041, P.R.C.)
Abstract:An image segmentation method is proposed which combines region growing algorithm with Pulse Coupled Neural Network (PCNN) model. In this method, the pixels of image are mapped onto the neurons in PCNN. The improved PCNN model's ignition frequency matrix and the regional growing theory are combined together to form this method. Experimental results show that the segmentation images in this method can keep more complete edge information and more accurate regional divisions, and the results are more in line with the recognition feature of human vision compared with some traditional segmentation methods.
Keywords:digital image processing  image segmentation  region growing  Pulse Coupled Neural Network
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号