Search

Image production tools do not always create a clear image, noisy and blurry images are sometimes created. Among these cases, Poissonian noise is one of the most famous noises that appear in medical images and images taken in astronomy. In recent years, various methods have been proposed to improve the quality of the image that has been lost due to this noise. For example, we can refer to methods that use fractional-order and second-order total variation priors or proximal thresholding. In this paper, in the first step, based on framelet transform, a local minimal prior is introduced, and in the next step, this tool together with fractional calculation is used for Poissonian blurred image deconvolution. The framelet transform domain of images usually have sparse representations. It is also well known that the use of framelet transfer has a proper effect on the edges of the restored image. Also, In this study, both blind and nonblind problems are considered. To evaluate the performance of the presented model, several images such as real images have been investigated. Various tools are used to study the efficiency of the proposed method such as PSNR and SSIM. The proposed method is compared with the existing methods such as fractional - order and second-order total variation. The simulation results show the appropriate representation of the proposed method in solving this type of problem. [ABSTRACT FROM AUTHOR]

One of the most important issues in image processing is the approximation of the image that has been lost due to the blurring process. These types of matters are divided into non-blind and blind problems. The second type of problem is more complex in terms of calculations than the first problem due to the unknown of original image and point spread function estimation. In the present paper, an algorithm based on coarse-to-fine iterative by l 0 - α l 1 regularization and framelet transform is introduced to approximate the spread function estimation. Framelet transfer improves the restored kernel due to the decomposition of the kernel to different frequencies. Also, in the proposed model, a fraction gradient operator is used instead of the ordinary gradient operator. The proposed method is investigated on different kinds of images such as text, face and natural. The output of the proposed method reflects the effectiveness of the proposed algorithm in restoring images from blind problems. [ABSTRACT FROM AUTHOR]

Restoring images corrupted by noise and blur is a burgeoning subject in image processing, and despite the large number of proposed restoration algorithms, the effort to bring about some improvement is always of great interest. The definition of fractional derivatives in recent years has created a powerful tool for this purpose. In the present paper, using fractional-order total variation and framelet transform, the nonconvex model for image restoration with impulse noise problem is improved. Then by alternating direction method of multipliers (ADMM) and primal-dual problem, the proposed model is solved. The convergence of the proposed algorithm is studied, and the proposed algorithm is evaluated using different types of tests. [ABSTRACT FROM AUTHOR]