Application Research Of Pixel Polarizer Array Imaging Technology In Image Enhancement

As a branch of artificial intelligence,machine vision technology has been developed for decades and is widely used in industrial detection,target recognition,satellite remote sensing,marine detection,and unmanned driving.Traditional machine vision technology obtains two-dimensional light intensity information of the target scene through photoelectric conversion.Detecting only a single intensity information has limitations in many application scenarios.When the detection environment is poor,the visibility of the field of view is low,and the background noise interference is large,the intensity images obtained by the traditional detection technology often have the disadvantages of blurring and low contrast,which cannot meet the task of target recognition and detection.Therefore,a new imaging technique that is simple,fast,stable,and highly real-time is needed to make up for the above disadvantages.Polarization imaging technology has unique advantages in the field of image enhancement.The latest polarization imaging device based on pixel polarizer arrays has the advantages of small size,stability,and high real-time performance.So,this paper studies the principle of imaging technology based on pixel polarizer array,and mainly studies its application in the field of image enhancement such as underwater target detection and image fusion.This topic mainly carries out the following three aspects of research work:(1)The structure and principle of the polarizing camera based on the pixel polarizer array are analyzed.In order to correct the systematic errors of the polarizing camera,a calibration experiment is performed on the polarizing imaging system with reference to the E-P calibration method,and 36 × 50 sets of polarization angle data are collected through experiments to jointly calculate the actual system matrix,and the accuracy of the system is verified by using known polarized light detection experiments.The experimental results show that the polarization angle detection error is within 0.5 degrees,which meets the needs of practical applications.In order to obtain four different polarization angle images of the detection target to obt ain polarization information,the original image was processed using an interpolation algorithm based on the micro-arrangement structure of the pixel polarizer,and the image splitting effect of bilinear interpolation algorithm and bicubic interpolation algorithm is compared.(2)A polarizing camera based on a pixel polarizer array was used to perform target detection experiments in a turbid water environment.Based on the underwater imaging physical model,a polarization descattering algorithm based on the Stokes vector was proposed.In experiments using coins as targets,a descattering algorithm was used to process four different polarization angle images of the target(0 °,45 °,90 °,135 °)to obtain the target's descattering image.Tests were performed at different imaging distances.The experimental results show that the method in this paper is more suitable for target-enhanced imaging in water bodies with mild and moderate attenuation,and the contrast can be increased up to 140%.At the same time,in order to verify the universality of the system method,the card,camouflage cloth,leaves,and plastic bottle caps were selected as underwater detection targets for experimental analysis.The experimental results show that the above objects all have a certain descattering effect.(3)The feasibility of image enhancement based on pixel polarizer array and image fusion algorithm was studied and the enhancement effect was evaluated.The imaging experiments were performed under the conditions of natural light illumination and polarized light illumination respectively.The discrete wavelet transform was used to decompose the intensity image and polarization image of the target scene into wavelet coefficients.A fusion rule based on local energy plus edge feature extraction was proposed for Image for fusion reconstruction.The experimental results show that the fused image is superior to the original intensity image in terms of gray average,standard deviation,information entropy,and contrast.The fused image ca n highlight the edge and texture details while retaining the basic information of the original image.At the same time,the effect of the fusion rule on local energy plus edge feature extraction and the traditional mean plus maximum value fusion rule are compared.Based on the above researches,this paper analyzes the structural principle of the imaging technology based on the pixel polarizer array,realizes the image enhancement of turbid underwater targets,and enhances the detailed information of the image by combining the polarization image fusion technology.These results make it possible to apply polarization imaging technology to more fields.