Nighttime foggy image generation algorithm based on semi-analytic model

To solve the problems that there were few studies on the generation of night foggy images, and the commonly used atmospheric scattering model was not effective in generating foggy nights, a new night foggy image generation method was proposed. First, since artificial light sources often dominate the night scene, a semi-analytical foggy imaging model that describes artificial light sources and halo phenomena was constructed. Then, an intrinsic image decomposition method was used to decompose the night image into illuminance map and reflection map to obtain the relevant parameters of the semi-analytical model. Finally, the atmospheric modulation function and the point spread function were used to further simulate the halo effect of the artificial light source, and the final generation effect can be obtained. The experimental results show that, compared with other methods, the proposed method can better simulate the halo effect of artificial light source under nighttime fog. Meanwhile, the statistical results of AuthESI indicators and volunteer preference experiments also reflect that the simulation effect of the proposed method is more realistic and natural compared with other methods.