Fast track article for IS&T International Symposium on Electronic Imaging 2020: Autonomous Vehicles and Machines proceedings.

No abstract

No abstract

Image haze removal is essential in autonomous driving as the outdoor images captured during unfavorable weather conditions, such as haze or snow, are affected by poor visibility. Much research has been done to overcome image degradation such as low contrast and faded color due to haze. However, in the traditional model, a phenomenon is neglected that several particles simultaneously involved in light acquisition. To address this problem, we propose a novel single image dehazing method based on the spatially adaptive atmospheric point spread function (APSF). We developed a module that estimates the APSF to overcome the limitations of the spatially invariant APSF which used in existing dehazing algorithms. The key factor in the estimation is that road scenes with haze have different statistical characteristic from common hazy images in color and resolution. Furthermore, the APSF on the traffic signs or lights is estimated by generating superpixels to prevent halo artifacts around the sharp edges of the images. We adopted the total variation model as a regularization functional to reduce halo and unnatural artifacts that may occur during deconvolution. The haze-free images from the proposed method tested whether the proposed method can enhance the performance of vision algorithms for autonomous driving. The experimental results demonstrate that the proposed method outperforms state-of-the-art image dehazing methods enhancing the performance of the vision algorithms. Moreover, additional experiments demonstrated the effectiveness of the proposed method for quantitative and qualitative comparison with the state-of-the-art algorithms.

Images acquired from the outdoors are often degraded owing to bad weather conditions, such as haze, fog, and rain. Various algorithms for dehazing daytime images have been introduced to eliminate the effects of haze. However, unlike daytime conditions, where sunlight is the main source of light, there are multiple regional light sources in nighttime conditions. Thus, it is difficult to effectively eliminate the effects of haze at night using dehazing algorithms that primarily target daytime hazy images. In addition, most nighttime dehazing algorithms adopt the dark channel prior (DCP) to estimate the transmission, but this approach has problems due to the spatially variant illuminations. In this paper, we propose a nighttime single image dehazing algorithm based on the structural patch decomposition. First, atmospheric light is estimated by separating the intensity and color of light. To estimate the transmission, the candidates for the transmission values are set, and the final transmission value is obtained based on a weighted summation of these candidates for each image pixel. The weights of the candidates are determined according to the structural patch information of the scene radiances estimated with these candidates. Using this approach, an appropriate transmission map can be obtained, and the dehazing procedure is shown to be robust to the mixed illumination conditions observed in the nighttime conditions. The experimental results show that the proposed algorithm effectively eliminate the effect of haze in images based on quantitative and qualitative evaluations.INDEX TERMS Nighttime dehazing, atmospheric scattering model, visibility restoration, contrast enhancement.