摘要
研究了一种将新特征提取方法(13维特征提取法)与量子神经网络相结合,来实现手写数字识别的方法。13维特征提取法是从每个字符中提取关键的13个特征值,而量子神经网络是将神经元与模糊理论相结合的模糊神经系统,能很好地减少模式识别的不确定度,提高模式识别的准确性。通过使用MNIST样本库仿真比较实验可知,该方法具有设计算法相对比较简单,且识别正确率较高的特点。
A new handwritten digital recognition method based on a 13-D feature extraction and quantum neural network （QNN） is presented. The 13-D feature extraction technique extracts the key 13 spots as characteristic point from each digits character. QNN is a neuro-fuzzy system merging neural modeling with fuzzy-theoretic concepts, it can reduce the uncertainty of pattern recognition and improve the veracity of pattern recognition. The experimentation results show that the method presented here is has the characteristic of simplity and high classification rate, by taking the MNIST database as examples.
出处
《电子测量技术》
2009年第6期84-87,共4页
Electronic Measurement Technology
基金
国家自然科学基金(50277010)
湖南省自然科学基金(07JJ6132
05GK2005)
教育部博士基金(20020532016)
高等学校新世纪优秀人才支持计划(NCET-04-0767)
关键词
手写数字识别
量子神经网络
特征提取
handwritten digital recognition
quantum neural network
feature extraction
