摘要
农村房地一体档案是对农村宅基地、集体建设用地使用权及房屋所有权进行确权登记的重要依据,将签章后的纸质档案转为电子档案进行存储对不动产权证书办理具有重要意义。由于目前缺乏能识别档案内容并进行分类归档的工具,设计并实现了基于Tesseract-OCR的农村房地一体归档系统。使用光学字符识别(Optical Character Recognition,OCR)对档案扫描图像进行识别,训练校正字库,提取图像中的文字信息,实现档案资料的分类存储。运用四川省某县的部分房地一体档案进行系统测验,应用结果表明,系统的识别归档准确率为96.5%,能满足房地一体档案归档需求,降低了人工识别归档的繁琐性,极大提高了归档的工作效率,提升了档案分类的准确度。
For the housing ownership and the use right to the rural residential land and the collective construction land,the archives of rural real estate are an important basis.For handling the real estate ownership certificate,it is of great significance to store the electronic archives instead of the signed paper.Due to the lack of tools to identify and classify archives,the study designs and implements the archiving system for rural real estate based on Tesseract-OCR.Optical Character Recognition(OCR)is used to recognize the scanned image,train and correct the character library,extract the text information in the image,and finally realize the classification and storage of the archive.Part of rural real estate archives from Sichuan Province is used for systematic testing.The application results show that the identification and archiving accuracy of the system is 96.5%,which can meet the archiving needs.The system can reduce the complexity of manual identification and archiving,and greatly improve the efficiency and the accuracy of archive classification.
作者
谭静
Tan Jing(Survey and Planning Research Center of Sichuan Geological S&R Institute,Chengdu,610017,China;Geological Surveying and Mapping Institute of Sichuan Province,Chengdu 610017,China)
出处
《黑龙江科学》
2024年第12期154-157,共4页
Heilongjiang Science
