Abstract:In order to improve the imaging quality of mine images and improve the observation effect, an improved Retinex mine image enhancement algorithm is proposed for the problems of color distortion, halo blur and over-enhancement when the traditional Retinex algorithm is used to process mine images. Firstly, the image to be processed was changed from RGB space to HSV space. Based on Retinex theory, the improved adaptive fast-guided filtering of V component was used to estimate the illuminance, and then the reflection component was obtained. An "S-type" function contrast component was proposed for illuminance equalization. Non-linear stretched of the reflected component for enhanced detail. Finally, the processed illuminance component and the reflection component were merged, and the RGB space was converted back to obtain the final enhanced image. The algorithm was applied to the non-uniform illumination environment under the mine, and three representative algorithms were selected for comparison. The experimental results show that the proposed algorithm is superior to other algorithms in subjective and objective evaluation. It is concluded that the algorithm is superior in color, detail and edge preservation, and can avoid over-enhancement and achieve effective enhancement of mine image.