Research On Spectral Polarization Image Correction And The Method Of Noise Reduction Based On AOTF

Spectral polarization imaging technology is a new technology in the field of photoelectric detection.Spectral polarization imaging is a combination of spectral technology,polarization technology and imaging technology.It can obtain polarization,spectral and spatial information of the observation target at the same time.It has great application prospect in military and civil aspects.Spectral polarization imaging uses infrared detector for image acquisition.The infrared imaging system has the advantages of simple structure,high sensitivity and low power consumption,so it has been highly valued and widely used.The typical feature of infrared image signal is that different image units correspond to different infrared detector units.Under the same external irradiance,the output response of each array element is inconsistent because of the nature of the device itself and the precision of the processing and the environment.It seriously affects the performance of the imaging system.Therefore,infrared image must be corrected nonuniformity.In this paper,the causes and forms of non-uniformity of infrared focal plane are discussed in detail,and the principle,advantages and disadvantages of several existing non-uniformity correction algorithms are analyzed.On the basis of the existing experiment with an s-shaped curve model of multi-point approximation nonuniformity correction algorithm,the algorithm compared two calibration more precise,relative to a multipoint line selection simple computational complexity.In this paper,the classical median filter method is selected for the blind element compensation process,and its window and calculation quantity are improved according to the principle of non-uniformity correction.Again the image noise reduction algorithm is studied,through the analysis of the noise and the comparison of the existing noise reduction algorithm,based on the advantage of the characteristics of wavelet denoising and the characteristics of wiener filtering are combined with improved obtained a better than other algorithm is an algorithm,to improve the signal-to-noise ratio by.The use of software on existing nonuniformity correction and noise reduction algorithm and the new algorithm is proposed in this paper the simulation comparison,good results have been obtained at the same time,the correctness of the theoretical analysis of the proposed algorithm is verified the effectiveness of the proposed scheme and.