Abstract: Infrared thermal imaging is a very promising method for evaluating coal and rock damage. To further process the infrared image and extract the key information, the damage status of coal and rock can be distinguished according to this quantitative information. With uniaxial loading, fracture development maps were developed, and synchronous acquisition of infrared images of the rock samples was carried out. The infrared images were analyzed and processed using a normalized histogram, and the details of the infrared images were quantitatively characterized. The results show that the gray value distribution of the infrared images at different times can reflect the surface temperature changes and stress values during the failure process of the sample. When the main fracture occurred, the percentage of pixels in the gray value interval 240, 255 (the surface temperature of the rock sample was 29.01℃-33.19℃) increased by 13.85% compared with that in the previous moment. In addition, the change trend of the proportion of pixels in the gray value interval 224, 255 over time is highly correlated with rock damage variables, which shows that the normalized histogram can characterize the damage and destruction process of the rock mass.