Improved retinex image enhancement algorithm based on bilateral filtering

In view of such problems as halo and color distortion for the traditional Retinex enhancement algorithm, an improved Retinex image enhancement algorithm based on bilateral filtering was proposed. In the HSI color space, firstly, the gamma correction was utilized to adjust the brightness of the image. And then, the image was decomposed into the illuminant image and the reflected image using the bilateral filter. Afterwards, the dynamic range of the illuminant image was compressed and the local contrast of the reflected image was enhanced. Finally, the saturation S was adjusted adaptively according to the correlation coefficient between the saturation S and the brightness I. The experimental results indicated that, the improved algorithm could not only suppress the halo and color distortion effectively, but maintain the color saturation and image authenticity, when was applied to color image. Thus, it was very conducive to the practical application of computer vision technology.