摘要
快速和准确地处理多元异构数据文档,是实现电子航海通告信息由互联网获取到生产实际应用的核心环节。基于对航海通告源数据结构与特点的分析,设计了通告预处理数据库模型、系统功能模型及工作流程,利用OCR技术、数据库技术及VS2010编程环境,研发了电子航海通告预处理系统,解决了通告文档内容的拆分与整合、识别与提取、检核与修编等关键技术问题。实验结果表明:该系统能够高效地提取通告信息,减少制图人员手工作业的工作量,基本满足编制航海通告产品、海图改正与更新等业务需求,为一体化海图生产提供高质量的通告数据源。
Fast and accurate processing of multi-source heterogeneous data documents is a central part of realizing the application of electronic notice to mariners(NM)obtained from the Internet into actual production.Based on the analysis of the structure and characteristics of NM source data,this paper designs a notice pre-processing database model,system function model and workflow,and develops an electronic NM pre-processing system by using OCR technique,database technique and VS2010 programming environment to solve the key technical issues such as splitting and integration,recognition and extraction,check and revision of notice document contents.The experimental result shows that the system can extract notice information efficiently,reduce the manual work of drafting staff,meet the basic business needs of NM product formulation,chart correction and update,and provide a high-quality notice data source for integrated chart production.
作者
吴婉婷
崔洪生
郭立新
朱书颖
WU Wanting;CUI Hongsheng;GUO Lixin;ZHU Shuying(College of Marine Sciences,Shanghai Ocean University,Shanghai 201306,China;Chart Information Center,Tianjin 300450,China)
出处
《海洋测绘》
CSCD
北大核心
2022年第1期78-82,共5页
Hydrographic Surveying and Charting
关键词
海图生产
海图改正
航海通告
识别文档
提取信息
chart production
chart correction
notice to mariners
document recognition
information extraction
