Title :
An improved image clearness algorithm based on dark channel prior
Author :
Huang Darong ; Fang Zhou ; Zhao Ling ; Chu Xiaoyan
Author_Institution :
Chongqing Jiaotong Univ., Chongqing, China
Abstract :
An improved image clearness method based on the theory of dark channel prior is proposed. Firstly, the single channel image coming from the minimum value of each pixel among the three color channels is computed; secondly, the atmospheric light is estimated by using the wavelet threshold method; thirdly, the transmission maps are calculated by using the atmospheric light, and then the image is recovered by using the atmospheric light. The experiment results show that this method can restore the fogging image effectively and reduce the time complexity which brought by Soft Matting algorithm greatly.
Keywords :
image colour analysis; image segmentation; wavelet transforms; Soft Matting algorithm; atmospheric light; color channels; dark channel; fogging image; image clearness algorithm; image clearness method; single channel image; time complexity; transmission maps; wavelet threshold method; Algorithm design and analysis; Atmospheric modeling; Atmospheric waves; Image color analysis; Image restoration; Scattering; Wavelet transforms; dark channel prior; defogging algorithm; wavelet de-noising method;
Conference_Titel :
Control Conference (CCC), 2014 33rd Chinese
Conference_Location :
Nanjing
DOI :
10.1109/ChiCC.2014.6896219
