Multi-channel Polarization Information Clarity Method For Haze Degraded Images

In haze environment,the image quality acquired by imaging equipment is seriously degraded.Low visibility results in inaccurate acquisition of image target feature information,which seriously reduces the perceived effectiveness of computer vision systems in military reconnaissance,civil aviation transportation and port logistics,such as missile inability to hit targets accurately in haze weather,frequent road traffic accidents,port closure and cargo ships unable to go to sea.Therefore,it is of great practical significance to study the method of image sharpening for haze degradation.In order to reduce the influence of heavy fog on the outdoor imaging equipment and to maximally improve the foggy image resolution,this paper introduces a haze degraded image clarity algorithm based on multi-channel polarization information.The main work is as follows:(1)Based on the theory of atmospheric physics,the polarization characteristics of light and the principle of polarized light detection are studied.The detection of polarization characteristics of light is transformed into the measurement of Stokes vector,and the detection results are presented in the form of mathematical expression.On this basis,polarization defogging model is derived,which provides a necessary solution for the accurate operation of the subsequent algorithm.(2)Based on the principle of optical Stokes vector detection,a multi-channel synchronous imaging polarization information detector is designed to obtain the polarization information of the target.Due to the manufacturing process of optical elements of multi-channel detector and the inconsistency of lens installation,the imaging parameters of different optical paths are different.In order to reduce the influence of this objective factor on the experiment,the preprocessing operations such as pixel distortion correction and image registration are carried out in this paper.(3)Based on the polarization fog removal model,an image clarity method based on polarization information is proposed in this paper.A method based on the comparison of normal distribution threshold is proposed to filter out the highlighted non-sky region in the image and accurately estimate the atmospheric light value at infinity,and the polarization degree of atmospheric light is accurately solved by filtering preprocessing method.The transformation from global quantification to local variables is realized.Finally,the algorithm proposed in this paper is compared with several typical defogging techniques.The authoritative and objective evaluation indexes in the field are used to analyze the data of the images before and after fog removal.The analysis shows that the algorithm proposed in this paper is better in clarity,contrast and other visual effects,and has a wider range of applications.