Design and Implementation of Fault Region Segmentation Method for Transformer Heat Sinks

Abstract

For the fault region identification of transformer heat sink, in order to accurately segment the abnormal region in the heat sink, this paper studies a new abnormal region segmentation method for transformer heat sink. Firstly, Gaussian filtering and 4-direction Laplace operator are used to preprocess the infrared image of the heat sink to get a cleaner image, which provides the basis for the subsequent fault region segmentation; then the FCM segmentation algorithm is used to pre-segment the image of the heat sink to remove the interference of the background, and to get the target region of the transformer heat sink with a rougher edge; finally, the region growth algorithm is used to finely divide the fault region of the heat sink to get a more accurate segmentation of the fault region of the transformer heat sink. Finally, the region growth algorithm is used to finely divide the faulty region of the heat sink to obtain more accurate different faulty regions of the transformer heat sink. After experimental verification, the segmentation method can accurately segment the different fault regions of the transformer heat sink, and has certain practical value.