离散正交S变换在人脸识别中的应用研究

Application Research of Discrete Orthogonal S-Transform in Face Recognition

人脸识别是模式识别领域的一个重要研究方向,有着广泛的应用前景.由于人脸图像受到光照、表情等因素的影响,在空间域直接使用线性方法具有一定的局限性,因此,从人脸特征的角度出发,提出一种新的人脸识别方法,即基于离散正交S变换(Discrete Orthonormal S-Transform,DOST)特征的人脸识别算法.首先,对人脸图像进行二维离散正交S变换得到人脸图像的特征,然后通过主成分分析对特征矩阵进行降维,提取特征向量,形成特征矩阵,最后用支持向量机分类识别.通过相应的实验证明本文提出的算法提高了识别的精确度和识别的速度.

Face recognition is an important research field of pattern recognition, it has a broad application prospects. Because the face image is affected by illumination, expression and other factors, it has certain limitations in the spatial domain directly using a linear method. This article propose a new facial feature extraction algorithm from the perspective of facial feature extraction. The new facial recognition algorithm is based on Discrete Orthogonal S-Transform facial feature extraction. First, face image feature is acquired by DOST face image. Then, Its dimension is reduced and eigenvectors are extracted by the method of PCA（ Principal Component Analy- sis ）, SVM （ Support Vector Machine ） is adopted to sort and distinguish. Experimental results indicate that the algorithm proposed to improve the accuracy and speed of facial recognition.