An improved color image defogging algorithm using dark channel model and enhancing saturation

摘要

Fog is an atmospheric phenomenon that significantly degrades the visibility of outdoor scenes. It is difficult to lock in and track off criminals clarity in the high-definition surveillance hazy image and driverless. In our research, we develop dark channel model, re-refined transmission map by transferring coefficient. Compared with other defogging methods based on single color image prior, our method is a effectiveness of the proposed defogging algorithm, enhanced the saturation of defogging image and a higher quality depth map can be obtained of fog removal. In this way, a clear hazy image is obtained while maintaining the fog removing quality. Moreover, a good quality defogging image can be validated of fog removal by PSNR and MSE.