摘要
针对角点特征检测进行人脸识别中特征配准度低、识别精度不高的问题,提出基于蚁群算法和支持向量机的人脸识别算法。首先采用支持向量机算法进行人脸多重特征检测提取,然后对提取到的特征信息采用蚁群算法进行训练分类,实现对人脸特征的准确检测和分类识别,最后在Matlab仿真平台上进行性能测试,仿真结果表明,采用该算法进行人脸识别的精度较高,训练过程的收敛性较好,计算开销降低。
Since the angular point feature detection for face recognition has the problems of low feature registration andrecog- nition accuracy, a face recognition algorithm based on ant colony algorithm and support vector machine is put forward. The sup- port vector machine algorithm is used to extract the face multi-feature. And then the ant colony algorithm is used to train and classify the extracted feature information to realize the accurate detection, classification and recognition of face feature. The per- formance is tested on Matlab simulation platform. The simulation results show that the algorithm has high face recognition preci- sion, good convergence in training process, and low computation cost.
作者
孙珊珊
SUN Shanshan(College of Information Engineering, Suihua University, Suihua 152000, China)
出处
《现代电子技术》
北大核心
2016年第21期92-94,98,共4页
Modern Electronics Technique
关键词
蚁群算法
人脸识别
支持向量机
特征检测
ant colony algorithm
face recognition
support vector machine
feature detection
