Abstract: The traditional method for detecting heating defects in power GIS(gas insulated switchgear) equipment mainly focuses on monitoring the far side signals of sensors, but its signal anti-interference ability is not strong, and the recognition false alarm rate is high. Therefore, an intelligent detection method for heating defects in power GIS equipment based on infrared image registration was proposed. The Harris corner detection algorithm was used to extract feature points from infrared images of power GIS equipment. By using the distance constrained feature point matching algorithm to roughly match feature points, a set of rough matching feature point pairs was obtained.And by using the random sampling consensus algorithm, the mismatched point pairs were eliminated in the set of feature point pairs, and the precise feature point matching was performed to complete the registration of infrared images of power GIS equipment. Then the preliminary identification of the overheated area in infrared image of registered power GIS equipment was carried out through threshold method. By using the relative temperature difference method, the temperature rise ratio of the overheated area was calculated, whether the overheated area is a heating defect was analyzed, and the intelligent detection of heating defects was achieve in power GIS equipment. The experimental results show that this method can effectively extract feature points from infrared images of power GIS equipment and eliminate mismatched feature points, obtaining registered infrared images of power GIS equipment. At the same time, it can effectively identify overheated areas, the maximum defect false detection rate is 3.22%, and the maximum missed detection rate is 4.35%, which has practical application value.