Font Size: a A A

Image Defogging Research Based On Dark Channel Prior

Posted on:2016-07-19Degree:MasterType:Thesis
Country:ChinaCandidate:X F ChenFull Text:PDF
GTID:2308330467473329Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Image defogging is a technology to enhance fogged images or to restore fog-free scenes, toshow the information of the image in a better way, so as to get better visual qualities of an imageor better subsequent analysis and treatment in computer system. In real life, defogging systemcan be widely used in many fields such as military, video monitoring, satellite imaging, and is ofgreat value and significance. Due to the complexity of the defogging, currently few algorithmcan solve the problem perfectly. The pursuit of faster and better defogging effect has alwaysbeen the goal of scientific research workers.This paper first introduces the atmospheric scattering model, which can be expressed as aphysical model and is widely used to explain the process of image degradation and formation ofa fogged image. Then is detailed analysis of defogging application of dark channel priorprinciple, analysis of the deficiencies in the processing speed and processing effect in the skyarea.Aimed at the heavy computation cost of image defogging based on dark channel priorprinciple, a new image defogging method is proposed. First the smaller image is gotten bydown-sampling the original image, then we get the smaller image’s transmission according to theoriginal defogging steps. Next, the transmission is adjusted to the original size and applied torestore the scene. This method is proved to not only retain excellent treatment effect of the darkchannel prior defogging algorithm, but also significantly enhance the processing speed of imagedefogging, especially for large size images.Aimed at the color distortion caused in process of defogging images with bright areas bydark channel prior, a new image defogging method based on dark channel with feedbackmechanism is proposed. By means of comparing images before and after defogged with darkchannel prior, color distortion regions are segmented by color difference, then the transmissionscorresponding to distortion regions are adjusted using gamma function, and finally the scene isrestored with adjusted transmission. The algorithm can efficiently alleviate the color distortioncaused in process of defogging on the image with bright areas using dark channel prior algorithm,and make the image defogged smoother and more natural.
Keywords/Search Tags:Defogging, Dark channel prior, Down-sampling, Feedback mechanism
PDF Full Text Request
Related items