基于自编码特征提取及弹性学习的手写数字识别

2014-07-19 17:43:57姜芳芳何明一王欣欣
现代电子技术 2014年10期

姜芳芳 何明一 王欣欣

摘 要: 针对自编码算法提取输入特征能更好地发现样本间的相关性的优点,以自编码算法提取待识别样本特征作为多层前向网络的输入,以弹性BP算法训练网络,并用MNIST手写数字数据库样本测试。从正确率、拒识率、错误率和可靠率4项性能指标方面与逐像素方法进行了综合对比测试。研究表明,采用自编码特征提取、多层前向神经网络作为分类器以及弹性BP 算法进行训练的手写数字识别,具有更快的收敛速度和更高的识别可靠率。

关键词: 多层前向神经网络; 自编码算法; 弹性BP算法; MNIST数据库

中图分类号: TN911?34; TP183 文献标识码: A 文章编号: 1004?373X(2014)10?0031?04

Abstract: As the autoencoder algorithm for input feature extraction is better in discovering the correlation between samples, an new approach is proposed for handwriting number recognition (HNR), in which the autoencoder algorithm is taken to extract the feature under recognition as the input of multilayer feedforward network, resilient back propagation (BP) algorithm is emplored to train the classifer, and some saples chosen from MNIST handwriting digits database are used to test the performance of this new approach. A comprehensive comparisons between this new approach and the pixel?by?pixel method is coducted in correct rate, rejection rate, error rate and reliability rate. This study results show that the proposed new approach (the autoencoder feature extraction, multilayer feedforward neural network classifier and resilient back propagation training algorithm are used together) has faster training speed and higher recognition reliability.

Keywords: multilayer feedforward neural network; autoencoder algorithm; resilient back propagation algorithm; MNIST database

手写数字识别(Handwriting Number Recognition,HNR) [1?2]主要指利用计算机智能的识别书写在纸张上的阿拉伯数字,是光学字符识别(Optical Character Recognition,OCR)技术的一个重要方面,有广泛的应用需求。多层前向神经网络[3](MultiLayer Feedforward Neural Network,MLFNN)由于具有很好的非线性映射能力,在非线性函数逼近、模式识别与分类等领域得到广泛的应用。在用神经网络进行模式识别的过程中,特征提取[4?5]是一个重要的环节,它会直接影响神经网络分类器识别的效果。基于神经网络的字符识别的输入特征提取一般采用逐像素方法[6],这是一种最简单的特征提取方法,其特点是算法简单,运算速度快,但对特征较少的图像,这种方法提取的信息量过大,适应性不强,不能很好的表示各类样本特征。本文在前人工作基础上,采用自编码算法[7]提取待分类样本的特征作为多层前向网络的输入,通过弹性BP算法[8]对网络进行训练,以期训练速度及识别效果的提升。……

登录APP查看全文