遗传算法在人脸识别中的应用研究

Application Study about Face Recognition Based on Optimization Genetic Algorithm

研究人脸图像识别准确率问题,人脸是一个非刚体,具有变形大,针对影响因素多且易受干扰,用传统的方法识别率低。为了提高人脸图像识别正确率,提出了利用遗传算法的人脸特征提取的识别方法。首先采用小波变换和张量主成分分析(PCA)方法对人脸图像进行特征提取,然后通过改进的遗传算法对PCA提取的特征进一步的优化,得到人脸最优人脸特征子集,最后根据最优特征进行识别。利用标准人脸识别库进行仿真,试验结果表明,相对其它特征提取的人脸识别方法,不仅具有识别速度加快,而且正确率高,是有效的人脸识别算法。

Human faces have large variation in shape at different time.Many influences,such as lighting,background,facial expressions and facial details,will easily affect the recognize results.This paper introduces the commonly used face recognition.Firstly,features of each face image are extracted by using the wavelet transformation and the tensor principal component analysis（PCA） algorithm.Secondly,a modified genetic algorithm is led to optimize the features extracted by PCA and obtain the optimal feature subset of face.The final identification is based on the optimal features.Experimental results on the standard face recognition database show that compared with other feature extraction for face recognition,the algorithm has higher recognition speed and accuracy,and is an effective face recognition algorithm.