基于改进神经网络的火车票识别算法的研究
投稿时间:2021-06-25  修订日期:2021-07-21  点此下载全文
引用本文:
摘要点击次数: 103
全文下载次数: 0
作者单位邮编
刘娴* 南京机电职业技术学院 453003
基金项目:江苏省自然科学(19JS01671);
中文摘要:为提高火车票识别精度和效率,将图像处理技术和BP神经网络结合,提出一种基于图像处理和BP神经网络的火车票识别算法。首先,通过图像预处理、目标区域的定位、二值化、倾斜校正和字符分割,提取火车票的身份证号码特征信息,建立特征信息库;之后,将特征信息库作为BP神经网络的输入,数字和字符类别作为BP神经网络的输出,建立BP神经网络的火车票识别模型。研究结果表明,与模板匹配和SVM相比,提出的方法可以有效提高火车票的识别精度和效率,识别精度高达97.7%,从而为火车票识别提供新的方法。
中文关键词:神经网络  二值化  字符分割  模板匹配  支持向量机
 
Research on train ticket recognition algorithm based on Improved Neural Network
Abstract:In order to improve the accuracy and efficiency of train ticket identification, an algorithm based on image processing and BP neural network is proposed by combining image processing technology with BP neural network. First, through image preprocessing, target location, binarization, skew correction and character segmentation, the identity card number feature information of train ticket is extracted and the feature information database is established, a train ticket recognition model based on BP neural network is established, in which the feature database is the input of BP neural network and the types of numbers and characters are the output of BP neural network. The results show that compared with template matching and Svm, the proposed method can effectively improve the accuracy and efficiency of train ticket recognition, and the recognition accuracy is up to 97.7% , thus providing a new method for train ticket recognition.
keywords:neural network  binary  character segmentation  template matching  support vector machine
查看全文   查看/发表评论   下载pdf阅读器