Abstract: A new infrared image enhancement method Heat-HE was proposed by combining the infrared thermal field distribution law with the histogram equalization image enhancement algorithm. Several images were selected by random sampling from the experimentally obtained infrared images, and they were processed by three methods of Heat-HE algorithm, grayscale world algorithm and automatic white balance algorithm.The results of the algorithm were judged by two evaluation indicators of average peak signal-to-noise ratio and average structure similarity. The results show that the average peak signal-to-noise ratio and the average structure similarity of Heat-HE algorithm are better than those of the traditional grayscale world algorithm and automatic white balance algorithm.Heat-HE algorithm makes the infrared thermal image of the clad layer have clearer defect contour based on retaining the original image features and can separate the clad layer defects from the background noise.