Single image dehazing by segmenting dark channel prior

다국어 초록 (Multilingual Abstract)

In image dehazing, the existing transmission estimators bring out the halo artifact at boundaries unless they adopt a refinement process with the high computational complexity. We analyze how the existing transmission estimation methods suffer from the halo artifact at the boundaries and observed that the elaborate, high computational refinement processes to remove the halo effect are excessive for dehazing. On the basis of the analysis and observation, we embed a simple segmentation logic in an existing transmission estimator, which is sufficiently accurate for dehazing . The experiment verifies that the proposed method significantly reduces the halo artifact without requiring any refinement process.