基于最小定义完备集实现复杂工程图数据提取

摘要

使用简单通用的描述定义完成复杂工程图数据提取是工程图数据管理中的关键技术.基于自由度理论分析了复杂工程图的数据组织结构,提出了描述复杂工程图数据的最小定义完备集.以直线表格化与文字表格化相结合改进了工程图表格识别技术,提高了工程图数据识别的准确性和完整性.结合最小定义完备集和工程图表格改进识别技术提出了能够处理复杂工程图数据提取的通用解决方案.运用Teigha技术开发的复杂工程图数据提取组件与PDM/PLM等管理系统集成,较好的完成了复杂工程图数据提取工作.%Engineering drawings data extraction with simple and universal semantic definition is one of key technologies in engineering drawings data management. Minimum well-defined sets that describes engineering data is provided, after analyzing the data structure of complex engineering drawings with degree-of-freedom theory. A optimized form recognition technology based on line-structured and text-structured is provided, the accuracy and integrity of engineering drawings data extraction is improved. A intelligent data extraction solution, that can handle complex engineering drawings data extraction, is provided based on the minimum well-defined sets. A data extraction module programed with Teigha is realized.