Real-time Defogging of Single Image of IoTs-based Surveillance Video Based on MAP

Abstract

Due to the atmospheric scattering phenomenon in fog weather, the current monitoring video image defogging method cannot estimate the fog density of the image. This paper proposes a real-time defogging algorithm for single images of IoTs surveillance video based on maximum a posteriori (MAP). Under the condition of single image sequence, the posterior probability of the high-resolution single image is set to the maximum, which improves the MAP design super-resolution image reconstruction. This paper introduces fuzzy classification to calculate atmospheric light intensity, and obtains a single image of IoTs surveillance video by the atmospheric dissipation function. The improved algorithm has the largest signal-to-noise ratio after defogging, and the maximum value is as high as 40.99 dB. The average time for defogging of 7 experimental surveillance video images is only 2.22 s, and the real-time performance is better. It can be concluded that the proposed algorithm has excellent defogging performance and strong applicability.