White Channel Prior: 개선된 Dark Channel Prior를 이용한 안개 제거 연구

Abstract

This paper proposes a new dehaze method to improve the contrast value and color intensity degradation of the result image that occurs when the haze is removed through Dark Channel Prior, a representative algorithm of the prior based dehaze method. Due to the estimate and restores by defined a dark channel based on the minimum values of the R, G, and B channels of the local patch, the existing dehaze method decrease the reconstruction image in contrast and color intensity. Therefore, to improve this, this paper proposes White Channel Prior that improve the reduced contrast and color intensity by reprocessing the image processed by the Dark Channel Prior algorithm in the opposite concept to DCP. The proposed method improves the image quality by reconstructing the contrast and color intensity degradation problem of the existing method. As a result of a comparison experiment between the proposed method and the conventional method, the dehaze image to which White Channel Prior method is applied shows a subjective quality improvement compared to the existing method results.