Abstract: A color fundus image enhancement algorithm is proposed, which it is considering the matching of hue preserving, brightness enhancement and detail enhancement to deal with the problems of poor contrast, low brightness, and color distortion after other image enhancement methods. Firstly, the fundus image in RGB space is converted to HSV space for adaptive power-law transformation based on brightness to solve the problem of brightness balance of fundus image. Secondly, the limited contrast adaptive histogram equalization method is used in Lab color space to improve the detail information of fundus image. Finally, experiments are carried out on Diaretdb0 fundus image database and compared with other enhancement algorithms. The experimental results show that this method has better visual effect than other image enhancement algorithms, improves low contrast and the color distortion of color fundus images. The underwater color image quality evaluation was increased by 37.6%, average gradient was increased by 54.23%, visual parameter measurement index was increased by 7%. It provides a new pre-processing method for the recognition, segmentation and classification of fundus images.