一种结合稀疏表示和切比雪夫矩的人脸识别算法

在基于稀疏表示的人脸识别算法的基础上,利用切比雪夫矩在图像重建及抗噪声方面的良好性能,提出了一种结合稀疏表示和切比雪夫矩的人脸识别算法,对有无加性噪声干扰的人脸图像进行识别.给出了详细的数学推导过程和算法实现步骤,并通过实验对算法进行了验证.针对扩展的Yale B人脸库和AR人脸库的识别结果表明,当特征空间维数为496时,该算法在不同光照条件和不同表情条件下的识别率分别为98.33%和88.72%,在添加椒盐噪声后像素破坏比例小于60%的条件下识别率为100%.与基于随机脸的最近邻分类法、最近子空间分类法及传统SRC算法相比,该算法在抵抗图像的细节信息变化方面具有更好的鲁棒性.

Face recognition algorithm by using sparse representation and Tchebichef moments

Based on the sparse representation-based classification(SRC) algorithm,taking the advantages of Tchebichef moments in image representation and noise robustness,a face recognition algorithm combining the sparse representation and Tchebichef moments is proposed to recognize the face images with and without additive noise.The mathematical derivation and the specific procedure of the algorithm are given,and the experiments are made to verify the algorithm.The experimental results obtained from the extended Yale B database and the AR database show that when the feature space dimension is 496,the recognition rate of the proposed algorithm is 98.33% and 88.72% under various lighting conditions and expressions,respectively.When the proportion of pixels damaged by salt-pepper noise is less than 60%,the recognition rate is 100%.In terms of the robustness against various image details,this algorithm outperforms the nearest neighbor method,the nearest subspace method and the conventional SRC algorithm all of which are based on the randomfaces.