基于LW-CharNet的纸制票据档案文本检测  

Text Detection of Paper Bill Archives Based on LW Charnet

作　　者：刘清[1] LIU Qing(Xi'an Polytechnic University,Xi'an 710048 China)

机构地区：[1]西安工程大学,陕西西安710048

出　　处：《自动化技术与应用》2023年第10期77-80,共4页Techniques of Automation and Applications

摘　　要：为实现针对票据图像文本内容的高精度检测,提出一套基于LW-CharNet的纸制票据档案文本检测算法。介绍该模型的整体结构、票据图像的预处理方法以及主要的数据处理环节,并将该模型应用于票据文本检测实验。实验研究发现,所提出的LW-CharNet模型算法能够较为准确地标识出票据文本所在区域,相比于其他主流文本检测算法在准确率方面体现出显著优势,具有一定的应用价值。In order to realize the high-precision detection of bill image text content,this research proposes a set of paper bill file text detection algorithm based on LW CharNet based on the general idea of text detection and the general characteristics of bill text.The overall structure of the model,the preprocessing method of bill image and the main data processing links are introduced in detail,and the model is applied to the bill text detection experiment.Through experimental research,it is found that the LW CharNet model algorithm proposed in this study can more accurately identify the region where the bill text is located.Compared with other mainstream text detection algorithms,it shows significant advantages in accuracy and has certain application value.

关 键 词：文本检测 LW-CharNet 损失函数 数据标注 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]