Classification and recognition method of aerial photo insulators based on improved K-means algorithm

Insulator is one of the most important infrastru ctures in transmission lines, and its accurate identification is an important prerequisite for self-monitoring and fault diagnosis of transmission lines. In order to accurately identify the insulators in UAV aerial photography inspection, a classification and identification method based on improved K-means algorithm was proposed. Firstly, the insulator image is processed gray-scale, and the improved K-means algorithm is used to cluster the gray-scale image. Secondly, the segmentation defect is compensated by morphological filtering. Finally, according to the mean color difference of the target area of the insulator, the classification problem of the insulator is simplified to a one-dimensional data classification problem, so as to achieve classification recognition. The experimental results show that the proposed method can quickly and accurately classify and identify the insulators under complex background and the same shooting Angle, and the total recognition rate can reach $\mathbf{9 5. 5 6 \%}$, which provides a new idea for the classification and recognition of transmission line insulators in UAV inspection.