摘要
为提高正确识别率,将Eigenfaces、EigenUpper、EigenTzone和二阶特征脸法四种方法的初步识别结果先模糊化,然后采用模糊综合函数加权组合后获取新的距离函数以得出最终的识别结果.通过针对ORL数据库的实验表明,应用多特征模糊加权算法进行人脸识别是行之有效的.
In order to improve correct recognition rate,the elementary recognition results of eigenfaces,EigenUpper,EigenTzone and two-order eigenfaces are firstly fuzzy fused.And then the fuzzy integration function is adopted to combine the fuzzy results in order to obtain new distance function and the final recognition results are given.The experiments based on ORL face database show that applying fuzzy multiple feature combination method is fairly efficient.
出处
《山东理工大学学报（自然科学版）》
CAS
2010年第4期29-31,共3页
Journal of Shandong University of Technology:Natural Science Edition
关键词
多特征
人脸识别
模糊加权
模糊综合函数
multiple feature
face recognition
fuzzy combination
fuzzy integration function
