An Improved Image Restoration Algorithm Based on the Criminisi Algorithm and Least Squares Method

Image restoration is a hot topic in the field of digital image processing. Criminisi algorithm is an image restoration algorithm based on texture synthesis. In this paper, the basic principle of Criminisi algorithm is analyzed, the limitations of the priority definition in the algorithm are analyzed, and a solution is proposed by combining least squares fitting. Experiments on artificially coded images show that the improved algorithm has better restoration effect.