The Application Of Particle Swarm Optimization Algorithm In Image Retrieval

With the rapid development of computer network and the image technology,the traditional text-based image retrieval technology has not qualified people increasingly complex retrieval,so,content-based image retrieval has emerged.However,the retrieved images cannot meet the needs of users.Therefore,relevance feedback technology was introduced into the content-based image retrieval,makes the image retrieval accuracy is greatly increased.In CBIR,relevant feedback is interactive retrieval for many times,first relevant feedback is built on a specific algorithm.The first retrieval is crucial for later retrieval,it makes the feedback has significant limitations.In order to make the algorithm more accurately,it will repeat again and again.The efficiency of the algorithm is reduced.In this paper,the research content mainly has the following several points:(1)Combined shuffled frog leading algorithm's grouping ideas with particle swarm optimization,applied to relevant feedback process,treat it as a process of using the particle swarm optimization;(2)Aimed at the limitations relevance feedback algorithm and the problem of low efficiency,to optimize the feedback process of image retrieval combined PSO's iterative optimization and be able to jump out of the local optimum,improve the efficiency and accuracy of image retrieval.(3)For fast convergence and prone to premature phenomenon of particle swarm optimization,proposed a small-world particle swarm optimization with a Rule of degree.