The classical dark channel theory algorithm has yielded good results in the processing of single fog image, but in some larger contrast regions, it appears image hue, brightness and saturation distortion problems to a certain degree, and also produces halo phenomenon. In the view of the above situation, through a lot of experiments, this paper has found some factors causing the halo phenomenon. The enhancement dark channel algorithm of color fog image based on the local segmentation is proposed. On the basis of the dark channel theory, first of all, the classic dark channel theory of mathematical model is modified, which is mainly to correct the brightness and saturation of image. Then, according to the local adaptive segmentation theory, it process the block of image, and overlap the local image. On the basis of the statistical rules, it obtains each pixel value from the segmentation processing, so as to obtain the local image. At last, using the dark channel theory, it achieves the enhanced fog image. Through the subjective observation and objective evaluation, the algorithm is better than the classic dark channel algorithm in the overall and details.
Enhancement dark channel algorithm of color fog image based on the local segmentation
Selected Papers from Conferences of the Photoelectronic Technology Committee of the Chinese Society of Astronautics 2014, Part II ; 2014 ; China,China
Proc. SPIE ; 9522
2015-04-13
Conference paper
Electronic Resource
English
Color image segmentation based on adaptive local thresholds
British Library Online Contents | 2005
|Local Color Image Segmentation Using Singular Value Decomposition
British Library Conference Proceedings | 1998
|Color image segmentation using a self-organizing map algorithm
British Library Online Contents | 2002
|Hill-manipulation: An effective algorithm for color image segmentation
British Library Online Contents | 2006
|