Research Status and Development Trend of Image Restoration Technology

Abstract

Abstract Defocus of the imaging system, relative motion between the equipment and the object or inherent defects of the equipment will lead to the degradation of image quality. Usually, image restoration is required before image processing, and the results of restoration technology affect the effect of image processing. In order to study the effect of image restoration technology, the image restoration technology was systematically sorted out. According to the different restoration models, the image restoration technology is divided into the method based on regularization and the method based on Kalman filter, and the two methods are summarized and explained respectively. By analyzing the restoration models established by the two methods and their solving process, it is found that the method based on regularization has advantages in retaining image information, and the method based on Kalman filter has advantages in terms of speed. The development trend of image restoration technology is analyzed from the summary of the two restoration methods, and suggestions and thoughts are put forward for the development of image restoration technology. In the future, image restoration technology should make use of deep learning algorithm, combined with the actual environment and industry characteristics, to achieve intelligent, practical and domain.