基于PSO训练SVM的人脸识别

粒子群优化算法(PSO)是一种进化算法,操作简单,参数少。与传统算法和遗传算法相比较,PSO算法在众多应用中表现了更好的性能。在通过对PSO算法训练支持向量机算法研究后,利用支持向量机在学习能力方面表现的良好性能,结合核主元分析特征提取方法,将其应用于人脸识别中,该方法在实验中表现了良好的识别性能,为人脸识别领域提供了一条新的识别途径。

Face Recognition Based on Support Vector Machine Trained with Particle Swarm Optimization

Particle Swarm Optimization (PSO for short) is a evolutionary algorithm with simple operations and few parameters. Compared with traditional algorithm and genetic algorithms, PSO algorithm shows better performances in varies applications. After the research of support vector machine trained with particle swarm optimization, it integrated with KPCA will be applied in the recognition of face images, utilizing the good performances of support vector machine in study, and we will find a new way for face recognition.