An effective method for fog-degraded traffic image enhancement

In this paper, an effective method for fog-degraded traffic image enhancement is proposed. Firstly, the fog-degraded image is segmented into blocks and low-rank decomposition is carried out for these blocks. Then the block with the minimal sparsity is selected for the local transfer function computation. And the global transfer function is derived from the super-resolution reconstruction based on the double-cubic interpolation. Finally, the enhanced image is obtained by deconvolution of the fog-degraded image and the global transfer function. Our proposed method is conducted based on the traffic videos obtained under the same view and angle. Moreover, our proposed method is compared with several state-of-the-art enhancement methods including notch filter, BM3D and Retinex Model. And the enhanced images are applied for vehicle tracking by the means of BWH mean shift. The experimental results illustrate that our proposed method can effectively eliminate the fog, preserve the useful information and achieve a better performance in terms of both information-entropy index and visual qualities.