Research
Research Divisions
Research Progress
Location: Home>Research>Research Progress
An improved image dehazing and enhancing method using dark channel prior
Author: Update times: 2015-11-20                          | Print | Close | Text Size: A A A

In fog and haze weather conditions, the outdoor visibility is greatly reduced by the atmospheric scattering. Images taken in this weather suffer from serious degradation. Image dehazing based on the dark channel prior(DCP) is considered to be an elegant solution due to its advantages of simple implementation and excellent performance of dehazing. However, as it is based on the assumption that the transmission is locally constant, the patch size will affects the quality of dehazed images. A large patch size leads to bright atmosphere but serious halo artifacts, while a small one can achieve nice dehazing results with little halo artifacts but dim atmosphere. To achieve a nice dehazing reslut with little halo artifacts and good brightness atmosphere, an improved dehazing method based on the DCP and the guided filter(GF) was proposed in this paper. Our method differs from previous ones in two aspects. First, we take a small patch size(rd= 1) to solve the dark channel(DC), which can achieves a better contrast recovery with little halo artifacts compared to a middle one(rd= 7), then we proposed a brightness enhancement method on the dehazed image to solve the problem of dim atmosphere. Second, in the step of transmission optimizing, we take several gray scale images rather than the color hazy image as the guidance images. The experimental results show that the proposed method can achieve rather good dehazing results, but with a relative simple implementation and a low time complexity.

This research was published in proceedings of the 2015 27th Chinese Control and Decision Conference, CCDC 2015

 
Copyright © 2003 - 2013. Shenyang Institute of Automation (SIA), Chinese Academy of Sciences
All rights reserved. Reproduction in whole or in part without permission is prohibited.
Phone: 86 24 23970012 Email: siamaster@sia.cn