Fast Image Fog Removal Based on Gray Image Guided Filtering

摘要

Under the condition of foggy weather, the videos and images obtained by the imaging devices which capture visible light will be severely degraded due to the low visibility of the scene, such as contrast reduction and color attenuation. The fog removal algorithm based on the dark channel prior now has yielded a great effect, but the algorithm has the characteristics of high time complexity and space complexity; thus, it does not have practicality. On the basis of the dark channel prior, we propose a fast method of image fog removal based on gray image guided filtering. Firstly, we estimate the atmospheric scattering model transmission through the dark channel information, and then adopt the gray image of the input mistily image to guide the transmission filtering to enable the optimization of rough transmission, namely, to maintain edges and smoothing regions; on this basis, the recovery image without fog can be obtained. Experiments demonstrate that the proposed algorithm in this chapter can effectively remove fog from a foggy image and increase the efficiency as a result.