Abstract: In view of the problem of uneven brightness and blurry and unclear lighting in surveillance video images caused by low brightness and serious dust in the underground environment, an algorithm to enhance the underground video image based on two-dimensional gamma function is proposed. The algorithm uses edge-preserving guided filtering to extract the light components, adjusts the light through the two-dimensional gamma function to achieve uneven illumination adjustment, and uses the restricted adaptive histogram equalization to adjust the contrast of the entire image. Thereby improving the clarity and amount of information of the surveillance video. Compared with the retinex and gamma algorithms, the proposed algorithm can combine the advantages of the two, and is superior to the above two algorithms in terms of information entropy, average gradient, and standard deviation. The results show that it can effectively improve the clarity, amount of information and contrast of the image, and at the same time alleviate the phenomenon of uneven lighting and dust blurred image in the underground, and improve the overall visual effect of the underground video surveillance.