Image Inpainting Algorithm Based On PSO And FOA

In recent years,image inpainting has been a hot research topic in the field of computer vision.It is widely used to repair damaged images and remove specific objects in images.Among the many image inpainting algorithms,Criminisi algorithm and its improved algorithm are the most popular algorithms in this field because of its advantages of easy implementation,fast restoration and good repair effect.This paper mainly represents three steps of the original Criminisi algorithm: determine the priority and search the best matching block and replication,which explain some shortcomings of the original algorithm.Based on the analysis and improvement,the particle swarm optimization algorithm and the fruit fly optimization algorithm were introduced into the algorithm,and we can get a better restoration effect.Firstly,this paper introduces the research background,the research status and the application prospect of the image restoration,and then introduces the Criminisi algorithm in an all-round way.Secondly,analyzes the shortcomings of the original method in the calculation of priority.In the original algorithm,the confidence and the data item are product relation,when the image is gradually restored,confidence will soon drop to 0,at present,the data item has no effect on the priority,and the calculation of the priority will be inaccurate.In order to eliminate this problem,we change their relation to be addition,so that they have no affect on each other.Because the curvature of the isolux line can reflect the local features of the image,we introduce it to the calculation of the right of priority,which leads to the calculation of the priority more accurate.In the calculation of the similarity measure function,we also introduce the curvature of the isolux line.The experimental results show that the improved algorithm has better restoration effect,and the PSNR value is higher than the original algorithm.Then,because the original algorithm used global search method to search the best matching,which may takes a long time when searching.We introduce the particle swarm optimization algorithm and fruit fly optimization algorithm to search the best matching block,and we compare the time of they use to search.The simulation results show that the two improved methods have improved the efficiency of the algorithm.Finally,summarize the work I do during the graduate study and the further research on the subject.