钻井现场指针式仪表读数自动识别方法研究  

Study on Automatic Identification Method of Pointer Instrument Reading in Drilling Field

作　　者：刘海龙 马波 张奇志[1,2] LIU Hailong;MA Bo;ZHANG Qizhi(College of Electronic Engineering,Xi'an Shiyou University,Xian 710065,China;Key Laboratory of Shanxi Province for Gas and Oil Well Logging Technology,Xi'an Shiyou University,Xian 710065,China)

机构地区：[1]西安石油大学电子工程学院,陕西西安710065 [2]西安石油大学陕西省油气井测控技术重点实验室,陕西西安710065

出　　处：《控制工程》2023年第9期1665-1672,共8页Control Engineering of China

基　　金：陕西省教育厅科研计划资助项目(20JS123)。

摘　　要：钻井数据采集过程中,由于现场安装、拉线等限制,传输稳定性不高,采集效果并不理想。引入了SSD+MoblieNet目标检测算法快速检测表盘位置并将表盘从复杂环境分离。使用Retinex特征增强算法并改进自适应Gamma增强使得复杂光在仪表上表现得更加自然,并在原有的图像去模糊技术上加入了模板匹配技术对表盘图像进行去抖动恢复,稳定有效地获取圆盘信息,进一步获取表盘指针角度信息。实验表明,改进后的方案能够在光照不均以及小幅度抖动下将误差控制在3%内,基本满足钻井现场指针仪表的测量要求,相对于传统方案,有着更强的实用性。In the process of drilling data acquisition,the transmission stability is not high,and the acquisition effect is not ideal due to the limitations of field installation and wire drawing.The SSD+MoblieNet target detection algorithm is introduced to quickly detect the dial position and separate the dial from the complex environment.Retinex feature enhancement algorithm and adaptive Gamma enhancement are used to make the complex light more natural on the instrument.The template matching technology is added to the original image deblurring technology to restore the on-site dial image,so as to obtain the disc information stably and effectively,the dial pointer angle information is obtained for the next step.The experimental results show that the scheme can control the error within 3%under the condition of uneven illumination and small amplitude jtter,and basically meet the measurement requirements of oil drilling field pointer instrument.Compared with the traditional scheme,the scheme has stronger practicability.

关 键 词：仪表识别 目标检测 RETINEX算法 图像去抖 HOUGH变换 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TH89[自动化与计算机技术—控制科学与工程] TP391.41[机械工程—仪器科学与技术]