Non-uniform illumination image enhancement based on high dynamic range

Aiming at the problem of insufficient enhancement of dark places or overexposure of light places in the enhancement processing of high dynamic non-uniform illumination images, a weighted image enhancement method with fog image as weights is proposed. This method can improve the brightness of the dark while ensuring that the light is not overexposed. Firstly, the dark channel image is calculated from the original image, and the average value of the image lager than the threshold value in the dark channel image is taken as the atmospheric light value. Then, the fog image is computed based on the double scattering model. Finally, the foggy image is used as weights to weight the original image and the foggy image, and the weights are related to the pixel. The results of enhancement of highly dynamic non-uniform illumination images show that the PSNR and SSIM of this algorithm are higher than other algorithms. The processed image is closer to the original image in the bright places and closer to the fogged image in the dark places, which effectively avoids the problem of under-enhancement in the dark place and overexposure in the bright places of high dynamic.