首页|基于云平台的电子图纸管理信息自动提取技术

基于云平台的电子图纸管理信息自动提取技术

扫码查看
电子图纸主要以文件夹方式存储,导致电子图纸管理与应用较为复杂,制约了电子图纸的发展步伐,为此提出基于云平台的电子图纸管理信息自动提取技术研究.深入分析电子图纸管理信息内容,包括标题栏信息与明细表信息,结合云平台功能,将电子图纸包含管理信息部分进行明确切分,采用霍夫变换算法检测并消除管理信息中的表格线,提取管理信息特征,即Gabor特征与Gradient特征,基于深度学习算法制定电子图纸管理信息识别与提取流程,执行流程即可实现电子图纸管理信息的自动提取.实验数据显示:相较于对比技术来看,应用提出技术获得的电子图纸切分精度最大值为90%,电子图纸管理信息提取完整度最大值为93.02%,充分证实了提出技术管理信息提取性能更佳.
Automatic extraction technology of electronic drawing management information based on cloud platform
Electronic drawings are mainly stored in folders,which leads to the complexity of electronic drawing management and application,and restricts the development pace of electronic drawings.Therefore,the research on automatic extraction technology of electronic drawing management information based on cloud platform is proposed.Deeply analyze the content of electronic drawing management information,including title block information and detail table information,clearly segment the part of electronic drawing including management information in combination with the function of cloud platform,use Hough transform algorithm to detect and e-liminate table lines in management information,extract management information features,namely Gabor features and gradient fea-tures,and formulate the identification and extraction process of electronic drawing management information based on deep learning al-gorithm,The automatic extraction of electronic drawing management information can be realized by executing the process.The experi-mental data show that compared with the comparison technology,the maximum accuracy of electronic drawing segmentation obtained by the proposed technology is 90%,and the maximum integrity of electronic drawing management information extraction is 93.02%,which fully proves that the proposed technology has better performance of management information extraction.

electronic drawingsinformation extractionmanagement informationcloud platformopen interfaceinformation positioning

黄文汉、李青、聂靓靓、胡冬阳、张娜

展开 >

南方电网调峰调频发电有限公司,广州 511400

电子图纸 信息提取 管理信息 云平台 开放接口 信息定位

中国南方电网科技项目

022200KK52190006

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(4)
  • 15