The Calibration Method Of Pixelated Polarization Camera

Polarization is one of the intrinsic and very important attributes of light.Polarization information divides the amount of information in an image from three variables(Light intensity,spectrum and space)into seven variables(Light intensity,spectrum,space,degree of polarization,polarization angle,polarization ellipsoid,and circular polarization of rotation direction).It provides more abundant elements for the study on imaging method.There are some differences in the polarization information of reflected light between different targets.Analyzing and detecting these differences can get some detailed features of the target.First of all,the development and application of polarization technology are introduced in detail in this paper including the application of polarization detection technology in the military field,biomedicine and atmospheric polarization mode.Then this paper introduces the polarization imaging method in detail.The principle of polarization imaging is to calculate the information of degree of polarization and phase angle of polarization on target by measuring the light intensity information of several polarization phase angles.Traditional polarization imaging methods include: time-sharing polarization imaging and simultaneous polarization imaging.Compared with time-sharing imaging,simultaneous imaging is more suitable for moving target imaging of airborne moving base.At the same time,the polarizing imaging method based on prism splitters needs prism to separate light,and the optical and mechanical structure is complex.In this paper,the method of simultaneous polarization imaging with focal plane is adopted.The optical element is pixel level micro-polarizer.The micro-polarizer imaging based on pixel delineation has the following advantages: First,the imaging is in simultaneous mode,which is suitable for airborne moving base moving target imaging.Secondly,the optical and mechanical system has simple structure,suitable for the load with strict requirement for volume power consumption.In order to solve the non-uniformity of polarizer array caused by the difference of the precision and transmittance of each pixel micro-polarizer unit,calibration is essential.Based on the traditional least square correction algorithm,a new least squares super-pixel correction method with angle constraint is proposed in this paper.The mathematical model of calibration optimization algorithm with angle constraint is established,and the scalar gain and dark compensation parameters of super-pixel are added.Finally,an integrating sphere calibration system with micro-polarizer as the core is built.The optical path is measured horizontally,including spectral measurement and unbiased measurement of the light source,software debugging and precise control of the precision rotary table in the optical path.Experimental results show that the proposed algorithm not only effectively calibrates the degree of polarization,but also calibrates the non-uniformity of amplitude in the super-pixel,compared with the high-precision least square method.The polarization accuracy is increased from 0.7 to 1,and the inhomogeneity decreases by more than 8 times.The accuracy of the reconstruction angle was also increased by 10%.The peak value of light intensity is exactly at its corresponding angle.After interpolation,the edge performance is more obvious,which reduces the noise generated in the interpolation process and improves imaging quality.