Fast Algorithm of Image Enhancement based on Multi-Scale Retinex

Abstract

In this paper, a fast image enhancement algorithm based on Multi-Scale Retinex in HSV color model is presented. The image enhancement using Multi-Scale Retinex in HSV model is computationally complex due to nonlinear color conversion. The proposed algorithm for calculation of RGB color components uses linear color dependencies from the V-component of HSV model, which allows to obtain a result similar to the one which uses a nonlinear processing in the HSV model. The new red, green and blue values of an image pixel are computed by multiplying a new brightness with the brightness to RGB component values ratio of the original image. It allows to reduce image processing time on 30-75% depending on the image size. In addition, the usage of Haar wavelet transform with brightness correction by MSR only in the low-frequency area was proposed in order to minimize the processing time of large images. The experimental results show that color image enhancement by the proposed algorithm with Haar wavelet transform is faster in 2–2.8 times. Detailed experiments were performed on 80 color images having non-uniform brightness, different algorithms based on Retinex technology are implemented and their performance is compared.