摘要
传统基于二值图像的框线特征提取信息的票据识别技术,其结果存在一定的差错率,针对此问题,提出了基于颜色分割的发票识别与验证系统。通过颜色分割和版式分析,将文字区域、二维码及图章等不同区域进行分割,分区域进行识别处理后,对发票关键信息进行比对验证。与传统的票据识别技术相比,该系统不仅可以完成发票自动识别,还可以结合二维码信息进行自我验证,提高了发票识别信息的准确性和完整性。
Aiming at the problem of a certain error rate in the result of the traditional bill recognition technology based on the frame line feature extraction information of the binary image,an invoice recognition and verification system based on color segmentation is proposed.Through color segmentation and layout analysis,the text area,two-dimensional code,and stamps are divided into different areas,and the key information of the invoice is compared and verified after the area is identified.Compared with the traditional bill recognition technologies,this system can not only complete automatic invoice recognition,but also self-verify in combination with two-dimensional code information,which improves the accuracy and completeness of invoice recognition information.
作者
王涛
Wang Tao(Department of Electronics and Information Engineering,Tongji Zhejiang College,Jiaxing 314000,China)
出处
《信息技术与网络安全》
2021年第5期80-85,共6页
Information Technology and Network Security
关键词
颜色分割
发票
二维码
自动识别
OCR
color segmentation
invoice
two-dimensional code
auto recognition
OCR
