Concrete-filled steel tube structures are widely used in large-scale building structures, and the detection of concrete density inside the steel tubes has increasingly received attention. However, due to the outer steel tube encasement in concrete-filled steel tubes, it is difficult to directly observe internal defects. The infrared thermal imaging method utilizes specific infrared waves to detect and convert the detection results into visual images. When infrared waves pass through the air pockets within the steel tube concrete, heat accumulates on the tube wall, allowing the internal voids in the steel tube concrete to be visually identified. In this paper, concrete-filled steel tube specimens with different types of defects were created, and infrared thermal imaging was used to identify internal defects in these tubes. The cooling rate at defective points is significantly greater than the cooling rate at healthy points on the same section. However, due to heating and other external conditions that cannot be completely consistent between different components, there are variations in the cooling rates of healthy points. Therefore, the ratio of cooling rates at different points on the same section is used for defect identification. Through experiments, it was found that when the ratio of the cooling rate at a certain point on the same section to the cooling rate at a known healthy point exceeds 1.1, the measuring point may have a void defect.