改进的二叉树支持向量机在人脸识别中的应用

针对人脸识别问题，提出一种基于奇异值分解特征提取和改进的二叉树支持向量机实现多分类的人脸识别方法。在使用改进的二叉树支持向量机对不同人脸图像的奇异特征向量进行分类时，先利用Mercer核，将输入空间非线性可分的训练样本映射到高维特征空间Hilbert中，使之线性可分，将类超球体半径分解成核心半径和最小半径，通过两者加权计算最终的类超球体决策半径，并以此半径大小为依据生成二叉树结构。在ORL人脸数据库的仿真结果表明，该算法能有效提高人脸识别性能，具有较高识别率。

Face Recognition Based On Improved Binary Tree Support Vector Machine

A new approach for face recognition based on wavelet,singular value decomposition and improved binary tree support vector machine(BT-SVM)is presented.BT-SVM is applied to classify singular value feature vector of face image.The input space is mapped to a high-dimensional feature space(Hilbert) applying Mercer kernel function and the data can become separable in feature space.The decision-radius which comes from the category-hypersphere-radius that is divided into kernel-radius and max-radius resolve the final BT’s structure.The experimental results on Yale data-bases demonstrate the high efficiency of the algorithm in runtime and correct localization rate.