• Our work proposed to use image processing technology to process the acquired infrared video. • The improved super green algorithm is proposed to feature analysis in the actual situation. • The results showed that the proposed algorithm had convictive validity. In this study, we propose using an unmanned aerial vehicle infrared thermal imager to obtain infrared video, and using image processing technology to process the acquired video. Moreover, based on research regarding traditional algorithms, an improved super green algorithm is proposed, and a feature analysis is conducted according to the actual situation. In this study, an effective identification model was designed for the most common types of faults, the most common electrical equipment components for research were collected, tests for analyzing the effectiveness of the algorithm were designed, and relevant data and images were recorded. The research shows that the proposed algorithm has validity, and can provide a theoretical reference for subsequent related research. Image, graphical abstract In order to solve the problem that people are difficult to reach, people have proposed to use helicopters to inspect transmission lines. However, at this time, the helicopter inspection of the transmission line was mostly based on visual inspection, and the test results were not guaranteed. In order to ensure the continuity and safety of users, traditional manual inspection methods have not been able to achieve the desired goals. This research mainly introduces fault image segmentation and morphological processing algorithms. Theoretical research and experimental tests are performed on various segmentation algorithms such as histogram threshold segmentation, iterative threshold segmentation, and maximum inter-class variance. Combining the two fault image features, an RGB component constraint segmentation algorithm is proposed. Morphological processing is performed on the image after segmentation to improve the quality of the binary image. Summarize the fault status judgment and fault area identification methods of power equipment. For the rust failure of the hardware, the rust failure is judged by marking the fault area after the division and counting the area ratio. For insulator faults, combined with the regular arrangement of insulators, a bilateral contour difference matching algorithm was proposed to determine the state of the insulators and realize fault detection of power equipment. Fig. 1 shows the color image of the rust fault and its R, G, and B channel grayscale images. Since there is a strong correlation between the three components in the RGB space, and the pixel color is determined by the three channels together, the processing of the gray image of the specific channel alone has no adaptability and accuracy. Therefore, this method is used less. Fig. 2 shows the original grayscale image of the insulator. Fig. 3 is the original grayscale histogram obtained by statistical drawing on the basis of Fig. 2. It is difficult to obtain valid information from Fig. 2 and Fig. 3, and the fault feature cannot be extracted. A fault judgment algorithm is designed for two fault types. For the rust corrosion of the metal fittings, the rust fault is judged by red marking the divided fault areas and counting the area ratio. In addition, for the insulation fault, combined with the regular arrangement law of the insulator, a bilateral contour differential matching algorithm is proposed to judge the state of the insulator. In addition, this paper designs experiments to improve the effectiveness of the algorithm. The research results show that the results of this study have certain validity and can be applied to fault identification of power equipment. [ABSTRACT FROM AUTHOR]