Machine Vision Based Transformer Insulation Pad Identification Detection
Shuo Huang, Yali Wang, Yongfei Xiao, Jie Sun · 2023
Addressing the issues of low efficiency, low accuracy, and high cost in traditional transformer insulation pad recognition and detection, A machine vision based contour size detection system for transformer insulation pads was studied and designed. The image of the pad is captured by an industrial camera, the image is analyzed and processed by the vision software OpenCV, the shape is recognized by an improved edge detection algorithm, the minimum outer contour is fitted by the Douglas-Peucker algorithm, and the conversion of the spatial coordinate system is completed by fast camera calibration and pixel conversion, and finally the outer contour recognition and the length and width of different models of transformer insulation pads are realized. The results are displayed through the MFC user interface. The experimental results show that the pad identification and detection system is stable and can guarantee a high accuracy rate and solve various problems in traditional measurement.