Underwater Image Enhancement Algorithm Based on Improved Pyramid Fusion Method

Abstract

Aiming at the problems of color distortion, low visibility and loss of details in underwater image, this paper proposes an underwater image enhancement algorithm based on improved pyramid fusion method. The proposed method combines the advantages of white-balancing and contrast limit adaptive histogram equalization. Then, the Laplacian contrast weight, local contrast weight, exposure weight and saliency weight are combined in different ways. Finally, the enhanced image is obtained by improved pyramid fusion. The experimental results show that the enhanced images of the proposed approach outperform the contrast limit adaptive histogram equalization, dark channel prior and multiscale retinex with color restoration. It can effectively reduce color distortion, enhance contrast, highlight details, and make underwater images have better visual effect.