Image Defogging Algorithm Based on Limited Light Value and Transmittance Correction

In this study, an image demist algorithm based on limited light value and transmittance correction is proposed. The aim of the study was to address the issues of color distortion in the demist image obtained by dark channel prior demist algorithm when the filtering window is small, error in the selection of introduction factor and calculation of the transmittance of the bright area, and weak anti-noise performance of the demist image. First, the upper threshold of atmospheric light value A was set. Second, the best introduction factor was obtained by establishing the corresponding relationship between the introduction factor and structural similarity. On the basis of introducing the tolerance mechanism, the transmission optimization method was further proposed. Finally, based on the proposed defogging algorithm, a Gaussian filtering algorithm was incorporated, and the brightness of the defogging image was adjusted to improve the visualization effect. The simulation results showed that the PSNR and SSIM values and entropy value of the image obtained by the proposed algorithm were 9.9964 dB, 8.57%, and 0.3732 higher than those before the improvement, respectively; thus, the effectiveness and superiority of the proposed algorithm were verified.