Application Of Polarization Imaging Technology In Scattering Medium Imaging

Imaging a target in a scattering medium is an important research topic in the field of current image processing technology.It is an important problem to be solved urgently in the fields of underwater detection,military reconnaissance in complex environments,and intelligent driving in foggy weather.Scattering medium imaging mainly refers to imaging technology in complex environments such as turbid water,haze,and the atmosphere.When the signal light reflected by the target passes through this type of scattering medium,scattering and absorption effects occur,the signal intensity becomes low,and forward scattering noise and backward scattering noise are generated.In the end,the image we received was an unclear image with blurred,low SNR and low contrast.The traditional optical imaging technology often cannot get the effect we need.Therefore,how to study high-quality imaging in a scattering medium environment is a subject of great research value.The focus of this paper is on the application of polarization imaging technology in scattering media.Introduced the basic principles and related concepts of polarization imaging,four types of polarization imaging systems and multiple polarization image processing techniques,and analyzed their flaws and applicable scenarios in detail.The turbid aqueous solution was used as the scattering medium to give the polarization The result of image processing.The core content of this thesis is based on the traditional polarization imaging technology,a new polarization image representation method is proposed: polarization observation formula and polarization observation vector representation.This method can greatly simplify the imaging process and image processing process in the polarization imaging process,and can effectively reduce the probability of generating experimental errors.In the third chapter,the basic principles of the method,the method of use,and the calculation method for obtaining the full polarization information of the target scene quickly and easily using this method are introduced in detail.Based on this representation method,the traditional polarization imaging processing method is improved,which greatly simplifies the calculation of the polarization imaging process.In Chapter4,we us e experimental data to demonstrate this method and verify the correctness of this method.In engineering research,there are several problems when we describe the polarizationcharacteristics of polarized light using Stokes Vector: the description of the polarization characteristics of light is not intuitive,the processing of polarized images is complicated,and the errors generated when obtaining polarized images cannot be determined and resolved.By constructing a mathematical model of polarized light,this article visually describes the polarization characteristics of polarized light,simplifies the process of polarized image processing,and uses the least square method to calculate the cosine polarization characterization parameters,which effectively reduces the error produced by the polarization imaging process on the result.The effect makes up for the problems of the Stokes vector representation.