基于支持向量机和遗传算法的人脸识别研究

Face recognition based on support vector machine and genetic algorithm

采用自适应遗传算法(AGA)优化筛选改进高斯核函数支持向量机(SVM)参数模型进行人脸特征分类。支持向量机的泛化性能主要取决于核函数类型和核函数参数及惩罚系数C,本文在传统高斯核函数基础上提出改进高斯核函数作为支持向量机的非线性映射函数,并使用自适应遗传算法优化筛选核函数参数和支持向量机惩罚系数,将优化后的SVM模型用于人脸库进行实验仿真。实验结果表明,本文方法比传统高斯核函数支持向量机分类器模型有更高识别率。

This paper uses the adaptive genetic algorithm （AGA） to optimize the support vector machine（SVM） model for face feature classification. The generalization performance of SVM mainly depends on the type of kernel function and kernel function parameters, this paper based on the traditional gaussian kernel function to improve the gaussian kernel function as the nonlinear mapping function of SVM, and uses adaptive genetic algorithm to optimize selection kernel function parameters and SVM punish coefficient, the optimized SVM model is used to face library experiment simulation. The experimental results show that the method than the traditional gaussian kernel function of support vector machine classifier model has a higher recognition rate.