计量学报2025，Vol.46Issue(9)：1395-1402,8.DOI:10.3969/j.issn.1000-1158.2025.09.21
基于人工智能的齿轮测量数据数字化提取技术
Digital Extraction Technology for Gear Measurement Data Based on Artificial Intelligence
摘要
Abstract
The acquisition and application of data is the cornerstone of digital transformation,this paper discusses the challenges associated with extracting measurement data from imaging files of gear measurement report.These files have the disadvantage of not being directly editable or replicable by machine,requiring manual recognition and recording,resulting in low efficiency and a high margin for error,making them difficult to meet requirements for data accessibility,interoperability,and reusability.Therefore,based on artificial intelligence technology,GearOCR data recognition and extraction model is proposed.By optimizing of pre-processing,dataset expansion,network model,and verification rules,the gear measurement content from closed images into editable data by machine is extracted sucessfully.After testing with 50 data reports,the overall accuracy of recognition for GearOCR reached 100%,markedly enhancing the efficiency and accuracy of data extraction and providing a potential technology support for subsequent machine interaction and application in big data.关键词
齿轮测量/数据提取/数字化转型/人工智能/OCR技术Key words
gear measurement/data extraction/digital transformation/artificial intelligence/OCR technology分类
通用工业技术引用本文复制引用
卢粲,林虎,张勇,杨国梁..基于人工智能的齿轮测量数据数字化提取技术[J].计量学报,2025,46(9):1395-1402,8.基金项目
国家重点研发计划(2021YFF0600103,2017YFF0204804) （2021YFF0600103,2017YFF0204804）
