<p>The harsh environments of underground mines pose significant challenges for acquiring clear images from operational conveyor belts, thereby hindering accurate foreign object detection and threatening mining safety. In this study, a novel image enhancement technique is proposed to improve the clarity and quality of the collected images. The approach begins by converting images from RGB to HSV color space. The illumination component (V) is then estimated and corrected via a modified weighted distribution adaptive gamma correction (AGCWD) Subsequently, adaptive BM3D (ABM3D) filtering is applied to reduce noise while preserving details. After reducing the color cast through color correction, the S-channel is adjusted based on the contrast stretching algorithm to enhance the image saturation and contrast, and the image is finally transferred back to RGB space. Finally, the MSRCR algorithm is integrated to perform global color correction. Experimental results demonstrate that the proposed method effectively enhances brightness, mitigates uneven illumination, and avoids color distortion, making it highly suitable for enhancing images of foreign objects on underground coal transportation belts. Meanwhile, our proposed method shows significant improvement in several performance metrics such as information entropy and average gradient, which is compared with adaptive histogram equalization with restricted contrast, MSR and MSRCR image enhancement algorithms. Specifically, the information entropy increased by 16.94%, 5.29%, and 8.55%, respectively, and the average gradient improved by 115.26%, 156.91%, and 1.26%, respectively. These enhancements confirm the method’s novelty and its strong potential for practical application in improving safety and automation in coal mining.</p>