基于最小二乘支持向量机的压印字符识别方法

Pressed Protuberant Character Recognition Based on Least Squares Support Vector Machines

将最小二乘支持向量机引入到小字符集压印字符识别中.首先介绍最小二乘支持向量机的基本原理和主要算法,然后在实验中采用最小二乘支持向量机训练软件,针对标牌上的压印字符的数字集进行仿真,同时与神经网络等其他分类方法进行比较.实验结果表明此方法的识别率较高,在小字符集识别中具有较强的实用性.

This paper presents an application of least squares support vector machines in small-set pressed protuberant character recognition. The theory and algorithms of least squares support vector machines are introduced. Least squares support vector machines are used to train the software in the experiment for simulation of labels＇ pressed protuberant characters, and compare with the results of neural network classification, et al. Experiment results show that the least squares support vector machines method has high recognition rate and is practical.