TY - JOUR AU - Chang, Chin-Chen AU - Chang, Chin-Chen AU - Way, Der-Lor AU - Shih, Zen-Chung PY - 2019/01/01 Y2 - 2024/03/28 TI - Improved Stereo Image Dehazing Approach JF - European Journal of Applied Sciences JA - EJAS VL - 6 IS - 6 SE - Articles DO - 10.14738/aivp.66.5497 UR - https://journals.scholarpublishing.org/index.php/AIVP/article/view/5497 SP - 01 AB - <p>Numerous image dehazing algorithms have been studied intensively. However, most dehazing algorithms operate on single images. These algorithms produce inconsistent results if they are used to dehaze stereo images iteratively. In this paper, we present a novel dehazing approach for stereo images based on cross bilateral filtering. In this approach, we simultaneously estimate scene depth and dehaze the stereo images. The proposed approach is based on the observation of depth cues in the stereo images. Depth cues are mainly used to avoid inconsistent results, and the cross bilateral filter is used to preserve shape details. The results demonstrate that the proposed approach can deliver superior results to those of previously published methods.</p> ER -