Research on Image Segmentation Algorithm and Performance of Power Insulator Based on Adaptive Region Growing

With the widespread application of power inspections, the problem of insulator segmentation in complex environments has become a current challenge. An insulator image segmentation method based on adaptive region growing and the adaptive Otsu algorithm is proposed. The 8 neighborhood pixels are used for region growth, and the segmentation results are obtained through morphological processing. Finally, the original segmented image, dynamic threshold segmentation, global threshold segmentation, and adaptive region growth are quantitatively analyzed. For the result of natural lighting image segmentation, the accuracy of adaptive region growth segmentation is improved by 14.23% for the original segmentation. For the results of infrared image segmentation, the accuracy of adaptive region growing segmentation is improved by 8.13% compared with the original segmentation. Experimental results show that adaptive region growth threshold segmentation can extract contour information more completely, which has certain advantages compared with traditional threshold segmentation. It provides an important basis for the study of insulator fault diagnosis and infrared insulator temperature field feature extraction.