点击次数：

论文类型：会议论文

发表时间：2016-05-06

收录刊物：EI、CPCI-S、Scopus

页面范围：505-510

关键字：dark channel prior; saturation; transmission map; color correction

摘要：With the increase of spreading distance, the light energy decays rapidly in water. Simultaneously, the light will reflect and deflect multiple times in its travel path. Underwater image typically suffers from poor visibility, low contrast and color distortion due to light scattering and attenuation. In this paper, we put forward a novel algorithm to restore the image based on dark channel prior. The background light is estimated by combining the dark channel prior with saturation map. Then, the transmission maps of different color channels are obtained based on the ratios of attenuation coefficients to recover underwater image. At last, a color correction is applied to balance the shifted color. The experiments show that the method in this paper obviously improves the poor visual quality of underwater images comparing with the existing techniques.