基于压缩传感的手写字符识别方法

摘要

A robust method for handwritten character recognition with noises based on compressive sensing was presented. The sparsest representation of the test character computed by l1-minimization had distinct class information; therefore, it is easy to classify the characters. The experimental results show that the proposed method is a noise robust technique.%基于新出现的压缩传感理论,提出了一种鲁棒的手写字符识别方法,能很好地对含有噪声的字符进行识别.该方法通过对测试字符进行稀疏表示,采用l1范数最小化算法求得最稀疏的系数解,所获得的系数具有明显的类别信息,从而易于对测试字符进行分类.实验结果表明,该方法具有很好的噪声鲁棒性.