超分辨率复原( super resolution miage reco

Super - Resolution (super resolution miage reconstruction) by a certain signal processing techniques, from a plurality of degradation, a low resolution image high resolution image acquiring method. Maximum posterior probability based on Bayes estimation (MAP) method and a map projection onto convex sets (the POCS) method is the most effective methods. POCS mapping method was first proposed by Startk et al., The a priori information is actually a restoration method of restoring an iteration process. The method uses a priori knowledge of the solution as constraints, it is limited to a closed convex set, solved by iterative method. Having a simple structure, the advantages of easy to solve. When the posterior probability density of the original image is known, MAP estimation method based on Bayes made a good application. Schu ltz et al MAP method to solve a video sequence of super-resolution image restoration problem, G ibbs prior model with HuberM arkov, the restored constraint problem into a tape having only one solution to the optimization problem. Maximum likelihood estimation (ML) estimate of MAP is a special case, Katasagge los, who use this method while an estimated sub-pixel displacement and image noise, ML estimation problem solving method with the greatest expectations, and achieved good recovery effect.

Super resolution recovery is a method of obtaining a high-resolution image from multiple degraded, low-resolution images using some kind of signal processing technique. The maximum post-probability (MAP) method and convex projection mapping (POCS) methods based on Bayes estimates are the two most efficient methods. The convex projection mapping method was first proposed by Startk et al. and is actually an iterative recovery method that introduces a priori information into the recovery process. The method uses prior knowledge as a constraint of the solution, limiting it to a closed convex set and solving it by iterative method. It has the advantages of simple structure and convenient solution. When the post-probability density of the original image is known, the MAP method based on Bayes estimation is well applied. Schu ltz et al. solved the hyper-resolution recovery problem of video sequence image series image using MAP method, and using HuberM arkov G ibbs a priori model to transform the recovery problem into a constraint optimization problem with a unique solution. Maximum Likeestimation (ML) is a special case of MAP estimation, Katasagge los et al. use this method to estimate both sub-pixel displacement and image noise, solve the M L estimation problem with the maximum expectation method, and achieve good recovery results.

Super resolution MIA reconstruction is a method to obtain a high-resolution image from multiple degraded and low-resolution images through a certain signal processing technology. The maximum a posteriori probability (map) method and the convex set projection mapping (POCS) method based on Bayes estimation are the most effective methods. The projection mapping method of convex sets was first proposed by startk et al. In fact, it is an iterative restoration method that introduces prior information into the restoration process. In this method, prior knowledge is used as the constraint of the solution, which is limited to a closed convex set, and iterative method is used to solve the problem. The utility model has the advantages of simple structure and convenient solution. When the posterior probability density of the original image is known, the map method based on Bayes estimation has a good application. Schultz et al. Used map method to solve the super-resolution restoration problem of video sequence images. Using hubermarkov GBS prior model, the restoration problem was transformed into a constrained optimization problem with unique solution. Maximum likelihood estimation (ML) is a special case of map estimation. Katasagge Los et al. Use this method to estimate sub-pixel displacement and image noise at the same time, and use the maximum expectation method to solve the problem of ML estimation, which has achieved good restoration effect.<br>