Saturated image deblurring method based on nonlinear degradation model

Abstract

The invention belongs to the technical field of image processing, and discloses a saturated image deblurring method and system based on a nonlinear degradation model, and the method comprises the steps: inputting a saturated blurred image or a common blurred image and a blurred kernel corresponding to the blurred image; constructing a nonlinear blurred image degradation model according to a saturated image degradation mechanism; constructing a saturated image nonlinear deconvolution framework according to the obtained degradation modeling and the maximum posterior probability framework; determining a priori item, and constructing a nonlinear energy functional model by adopting total variation priori; and solving the nonlinear energy functional through an exchange direction multiplier algorithm, namely an ADMM algorithm or a split Bregman algorithm to obtain a clear image x to be restored. According to the method, the saturated nonlinear degradation characteristic is modeled, blurring in the saturated image is removed, the ringing phenomenon caused by errors of saturated pixels is effectively restrained, and a high-quality clear image is obtained through restoration.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a saturated image deblurring method based on a nonlinear degradation model. Background technique [0002] Currently, the closest state-of-the-art: Images captured in low-light conditions are often degraded by saturation blur. Imaging devices, such as smart phones, can perceive brightness with a dynamic range, and when the brightness range in the scene exceeds the dynamic range, the image will be saturated, and the brightness beyond the range will be truncated to the maximum value of the dynamic range, such as When the dynamic range is [0, 255], when the brightness exceeds this range, the corresponding pixel gray value is truncated to 255. [0003] The traditional deblurring algorithm is based on the linear degradation model, as shown in the following formula: [0004] y=Kx+n [0005] Among them, y is the input blurred image, K represents the matrix form of the blur ke...

AI Technical Summary

Problems solved by technology

[0012] To sum up, the problems existing in the existing technology are: the existing first-class saturated image deblurring algorithm is based on the saturation detection method, which is highly dependent on the accuracy of detection, and the accumulated detection error will directly affect the image restoration effect

The second type of method has a poor recovery effect when the nonlinear function is selected incorrectly or the parameters of the function are not selected correctly.

The difficulty in solving the above technical problems lies in the fact that the nonlinear truncation operator in the constructed saturated and blurred image degradation model is an unsmooth and unbreakable function, and it is difficult to model it into the existing maximum a posteriori probability framework MAP to achieve optimal restoration

The present invention proposes a smooth second-order derivable function that approximates the truncation operator to replace the truncation operator, but the approximate truncation operator is still a nonlinear operation operator, which brings challenges to the optimization process

Images