基于模糊度参数估计的数字媒体图像无损复原.

Traditional digital image restoration methods produce errors in the relative displacement between the camera and the subject during the actual shooting exposure time, resulting in poor restoration results. A lossless restoration method based on ambiguity parameter estimation is proposed. An ambiguity image degradation model is built to measure the relative motion direction and horizontal direction angle in the spectrum, estimate the motion ambiguity direction, and the motion ambiguity parameter by using the space Cartesian coordinate system. Through the radial basis function neural network structure, the input function is calculated, and the relative displacement of the camera and the shooting object during the actual shooting exposure time is compared with the estimated result, and the ambiguity error is calculated. Wiener filter restoration method is used to design the restoration process and realize the lossless restoration of digital media images. According to the experimental results, the fault tolerance of ambiguity direction of the proposed method is controlled within 1°, the fault tolerance of ambiguity degree is controlled within 0.2, and the improvement percentage of ambiguity degree is more than 40%. The restored images are relatively clear. [ABSTRACT FROM AUTHOR]