The Research Based On The Restoration Approach Of Super-Resolution Distilled From Highway Video Monitoring

ITS, which is widely utilized on the system of urban traffic intelligent control and expressway intelligent control, has became more and more popular in people's daily lifes since the early age of 21st century. The identification of license plate acts as a really essential part among all the functions of the vehicle-checking which is assessed to be indispensable in the system of ITS,in which we can get the conclusion that the license plate recognition has occupied as an crucial factor in traffic management.Since the running vehicles always keep a long distance that beyond the capture capability of surveillance camera,plus the factor of bad weather, the images recorded by the vedio can't satisty the request of license plate recognition; therefore,if we want to get a better image of license plate recognition, we should start with lifting the quality of image,and then identify the vehicles with a high graphics.Super-resolution restoration of video image is a good approach,this technique is to estimate a high resolution image from multiple low resolution observations degraded by warping, blurting, noise, and aliasing, remove the noise and the blurs caused by sensors and optics, and make the restored image break through its original resolution limitation.This paper is based on the approach of super-resolution restoration of video image with the aim of seeking a solution to solve the problem of hard-identification of blured image in highway surveillance video, and make a deep research of super-resolution restoration in license plate.This paper contains contents as below:illustrate the researching background of license plate recognition,illustrate the significance of the research in this paper,illustrate the develpment history and technich classification of super-resolution of video image,analyse the defect and shortage of the current methods,on the basis of a non-uniform sampling based on interpolation principle and to the iterative projection algorithm that combines the new algorithm, first through non-uniform sampling has been the principle of iterative interpolation factor, and then into the projector to the iterative algorithm. Analysis and comparison is made on the two common motion estimation algorithms, optical flow and block matching, as well as their application situations and limitations.