一种用于海上平台缆控注水井的多类型指针式仪表识别方法

A Method of Multi-type Pointer Instrument Recognition for Cable Controlled Water Injection Well on the Offshore Platform

通过摄像头对海上平台缆控注水井现场巡检时,需要能够自动读取仪表数据。为了实现不同类型指针式仪表的识别,提出了一种用于海上平台缆控注水井的多类型指针式仪表识别方法。首先,构建模板图像数据库,对每个模板图像进行预处理并绑定其参数信息,使用改进的YOLOv5模型检测待测仪表图像的模板类型,读取模板图像和其对应的参数信息。然后,进行模板匹配定位仪表位置,获取指针角度并计算仪表数值。实验结果表明,本文所提出的方法能够识别不同类型的指针式仪表,识别效果显著,定位表盘位置准确,指针角度和读数精度更高。

It is necessary to be able to read the instrument data automatically,when the cable controlled water injection well on the offshore platform through the camera is inspected.In order to realize the recognition of different types of pointer instruments,a multi-type pointer instrument recognition method for cable controlled water injection wells on the offshore platform is proposed.Firstly,the template image database is constructed,each template image is preprocessed and binds to its parameter information,the template type of the instrument image to be tested is detected by using the improved YOLOv5 model,and the template image and its corresponding parameter information are read.Then,template matching is carried out to locate the instrument position,obtain the pointer angle and calculate the instrument value.The experimental results show that the method proposed in this paper can identify different types of pointer instruments,and the recognition effect for pointer instruments is remarkable.The dial positioning is accurate,and the accuracy of pointer angle and instrument reading are higher.