基于粒子群优化算法的LS-SVM字符识别模型

LS-SVM character recognition model based on particle swarm optimization

提出一种基于粒子群优化算法优化相关参数的最小二乘支持向量机(LS-SVM)的字符识别模型。利用相关的字符数据,分别使用本方法和基于网格搜索的最小二乘支持向量机方法进行识别。仿真结果表明,该方法的精确度高于其它两种方法。

A character recognition model based on Least Squares Support Vector Machines（LS-SVM） is proposed in this paper,of which related parameters are optimized using Particle Swarm Optimization（PSO）.A case study based on character data is carried out using the proposed method,and general LS-SVM method.The result shows that LS-SVM with parameters optimized by PSO is more effective than the other.