Gray projection for single image dehazing
Wencheng Wang, Tao Ji, Xiaojin Wu, Feng Lin · 2018
Restoration of hazy image is an important issue in outdoor vision system. To accelerate the processing speed, a fast single image dehazing method is proposed based on the atmospheric scattering model and gray projection. In this method the transmission map is estimated roughly using minimum filtering which is based on dark channel prior and refined with fast average filtering. The gray projection method is adopted to obtain the atmospheric light. The subsection mechanism is designed to avoid the high light of sky region in recovered image, and Weber-Fechner Law is employed for color compensation in restored image. Our experimental results show that this algorithm can restore satisfied haze-free image, and keep the balance of quality and speed in real-time system.