摘要
提出一种新的基于变尺度的混沌遗传算法。该算法利用Logistic映射构成混沌序列,与遗传算法相结合,加快了种群的进化速度,并且在优化过程中具有较高的搜索精度和搜索效率。同时,给出了应用此改进的遗传算法训练支持向量机(SVM)参数的方法,将其应用在人脸识别中。理论分析和实验仿真表明,改进的GA-SVM方法比基本的GA-SVM方法能获得更好的识别效果。
This paper presents a new variable scalechaos genetic algorithm. The algorithm uses the Logistic map to build a chaotic sequence, and combines with genetic algorithm to accelerate the evolution speed of population, and also it has higher search accuracy and search efficiency in the optimization process. At the same time, it applies this improved genetic algorithm to train the method of support vector machine （SVM） parameters, then uses it in the face recognition. The theoretical analysis and simulation show that the improved GA-SVM method can obtain better recognition results than the basic GA-SVM method.
出处
《山西电子技术》
2013年第1期3-5,8,共4页
Shanxi Electronic Technology
关键词
遗传算法
支持向量机
变尺度
人脸识别
genetic algorithm
support vector machine
variable-scale
face recognition
