首页 | 本学科首页   官方微博 | 高级检索  
     

基于BP网络的人脸图像边缘检测算法
引用本文:赵宝柱,靳艳峰,耿科明. 基于BP网络的人脸图像边缘检测算法[J]. 计算机工程与设计, 2007, 28(21): 5181-5182,5260
作者姓名:赵宝柱  靳艳峰  耿科明
作者单位:中国邮政集团商函研究中心,河北,石家庄,050021;中国邮政集团商函研究中心,河北,石家庄,050021;兰州理工大学,计算机与通信学院,甘肃,兰州,730050
摘    要:采用改进的BP算法,实现人脸图像的边缘检测.构造了学习样本,并在较短时间内训练得到权值矩阵,从而实现二值图像边缘检测.在此基础上,综合灰度图像各位面的边缘提取结果,实现对灰度图像的边缘检测.利用ORL人脸数据库图像的实验,证明了将得到的权值矩阵用于人脸图像的边缘检测,泛化性较好,抗噪能力强,能得到较为连续精细的边缘.

关 键 词:反向传播算法  边缘检测  人脸识别  神经网络  二值图像
文章编号:1000-7024(2007)21-5181-02
修稿时间:2007-03-12

Edge detection method of face image based on BP neural network
ZHAO Bao-zhu,JIN Yan-feng,GENG Ke-ming. Edge detection method of face image based on BP neural network[J]. Computer Engineering and Design, 2007, 28(21): 5181-5182,5260
Authors:ZHAO Bao-zhu  JIN Yan-feng  GENG Ke-ming
Affiliation:1. Direct Mail Research and Consulting Center, China Post Group, Shijiazhuang 050021, China; 2. College of Computer and Communication, Lanzhou University of Technology, Lanzhou 730050, China
Abstract:An improved BP algorithm is proposed for edge detection of face images.Firstly,the samples are constructed and the weighted matrix is got by exemplifying in a short time span.Then the edge of binary images can be detected through the process.Furthermore, first divide each gray-scale image into 8 binary planes with different gray level,and then naturally the images surface up through synthe- sizing the edge of each binary plane.Applying the method to face images' edge detection,free continuous edge is acquired,the virtues of which lies in its strong resistance against noise pollution and is widely adopted in practice.
Keywords:BP algorithm  edge detection  face recognition  neural network  binary image
本文献已被 CNKI 维普 万方数据 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