An Image enhancement algorithm based on Gaussian weighted bilateral filtering and retinex theory

Abstract

An image enhancement algorithm based on Gaussian weighted bilateral filtering and Retinex theory is proposed to solve the disadvantages about halo, grey and noise amplification phenomenons of single-scale Retinex algorithm. First, the illumination images are estimated with Gaussian weighted bilateral filtering and their influence on visual effect is removed, which can overcome the halo phenomenon effectively. Then the contrast in low contrast regions is enhanced with piecewise linear transformation, which can improve the grey phenomenon of enhanced images. Finally, threshold values are determined by Otsu algorithm to identify the dark areas of images, and the noise in dark areas is removed by bilateral filtering. Experiments show that the new algorithm can remove halo phenomenon, enhance the contrast of images, get more abundant image information and overcome the noise amplification phenomenon, which is better than the conventional single-scale Retinex algorithm and the Retinex algorithm based on bilateral filtering.