A method and system for correcting infrared image non-uniformity based on averaging processing

Abstract

The invention discloses a method and system for correcting infrared image non-uniformity based on averaging processing, which belongs to the field of image processing and includes: using a one-way total variation model to obtain an estimated image u of the image to be corrected k ; Obtain the mean value of the pixel offset in each row when the gain is 1, and use it as the first offset estimate when the gain in the corresponding row is 1; Estimate the gain of each pixel, and use the mean value of the pixel gain in each row as the corresponding row. Gain estimate; obtain the difference between the offset estimate corresponding to the gain estimate and the first offset estimate to obtain the second offset estimate corresponding to the gain estimate; according to the gain estimate and the second offset estimate Correct the image to be corrected with the value to obtain the estimated image u″ k ;If the iteration termination condition is met, the image u″ will be estimated k as the final correction result; otherwise, update the iteration parameters to start the next iteration. The invention can suppress the non-uniform strip noise of the infrared image and preserve the edge details of the image as much as possible.

Description

technical field [0001] The invention belongs to the field of image processing, and more specifically relates to a method and system for correcting infrared image non-uniformity based on averaging processing. Background technique [0002] Linear Scanning Infrared Imaging System (LSIR) plays an important role in both military and civilian fields. Its imaging model is g i,j =a i × u i,j +b i , where i and j represent the row number and column number of pixels in the image respectively, u and g represent the observed infrared image and the final image output by the imaging system, u i,j and g i,j Respectively represent the pixels located in the i-th row and j-th column in the corresponding image, a i and b i Denote the gain coefficient and offset of the i-th row imaging detector response characteristics, respectively. Due to the inhomogeneity of the response of each detector, that is, the difference in the gain and bias of the detector, the infrared image contains a fixe...

AI Technical Summary

Problems solved by technology

[0005] The one-way total variational model can effectively suppress the non-uniform band noise in the infrared image, but because it does not consider the constraints of the imaging model in the linear infrared scanning imaging system, it makes the non-uniform band noise in the infrared image suppressed at the same time Image detail along the scan direction is also removed, blurring the edges of the image

Images