摘要
利用简化脉冲耦合神经网络(S-PCNN),提出一种处理椒盐噪声污染的人脸识别新方法.首先采用S-PCNN的相似群神经元同步发放脉冲特性对原图像进行噪声检测,然后结合数学形态学实现对噪声点的消除,最后使用S-PCNN的时间序列(OTS)和欧氏距离进行人脸识别.通过计算机仿真实验表明所提算法是有效的.
A face recognition method for dealing with salt and pepper noise pollution using Simplified Pulse Coupled Neural Network (S-PCNN) was proposed.Firstly the paper uses similar group of S-PCNN neurons issuing synchronous pulses to detect noise of the original image, and then combines with mathematical morphology to achieve elimination noise point,finally adopts the oscillation time sequences (OTS) of S-PCNN and Euclidean distance to process face recognition.Computer simulation results show that the proposed algorithm is effective.
出处
《云南大学学报(自然科学版)》
CAS
CSCD
北大核心
2015年第5期687-694,共8页
Journal of Yunnan University(Natural Sciences Edition)
基金
国家自然科学基金(61365001
61463052)
云南省应用基础研究计划(2012FD003)
关键词
S-PCNN
椒盐噪声
数学形态学
振荡时间序列
人脸识别
Simplified Pulse Coupled Neural Network (S-PCNN)
salt and pepper noise
mathematical morphology
oscillation time sequence
face recognition