基于YOLO算法的指针式仪表智能识别系统设计  

Design of intelligent identification system of pointer instrument based on Yolo algorithm

作　　者：孙怀远 张水利[1] 郑可心 SUN Huaiyuan;ZHANG Shuili;ZHENG Kexin(School of Physics and Electronic Information,Yan’an University,Yan’an 716000,China)

机构地区：[1]延安大学物理与电子信息学院,陕西延安716000

出　　处：《延安大学学报（自然科学版）》2022年第3期70-75,共6页Journal of Yan'an University：Natural Science Edition

基　　金：国家自然科学基金项目(61761042);陕西省教育厅科研计划项目(20JK0983);陕西省大学生创新计划项目(S202010719089)。

摘　　要：针对指针式仪表自动检测读数时,受复杂背景环境等因素影响造成仪表图像检测难、提取速度慢、仪表盘定位扭曲、倾斜等问题,提出基于YOLO算法的指针式仪表智能识别系统。首先,利用YOLO算法对工业指针式表盘进行精确定位,并使用透视变换算法进一步校正,克服表盘扭曲或倾斜带来的误差;然后,对得到的二值化图像使用几何矩算法遍历得到仪表表盘的圆心和半径,提取仪表的刻度区域和指针;最后,采用角度法计算仪表读数。实验结果表明,该系统的仪表读数识别准确率不低于95%,可满足工业应用的精确度要求。In view of difficult detection of instrument image,show extraction speed,distorted and tilted instrument panel positioning,tilted and other problems affected by the complex background environment when the pointer instrument automatically detects the reading,this paper proposes an intelligent pointer instrument recognition system based on YOLO algorithm.Firstly,YOLO algorithm is used to accurately locate the industrial pointer dial,and perspective transformation algorithm is used for further correction to overcome the dial distortion or tilt.Then,the center and radius of the dial are obtained by using the geometric moment algorithm,and the calibration area and pointer of the dial are extracted.Finally,the angle measure technique is used to calculate the meter reading.Results show that the accuracy of the system is not lower than 95%,sufficient for industrial applications.

关 键 词：仪表智能识别 YOLO 目标检测 

分 类 号：TH703[机械工程—仪器科学与技术] TP391.4[机械工程—精密仪器及机械]