Research of image retrieval algorithm based on PSO and a new sub-block idea

Image retrieval is a hot technology in the field of computer technology nowdays. This paper presents a novel image retrieval algorithm which is mainly used to solve problems for image retrieval in large image database. Firstly, preprocess the sample images, abstract the color histogram information of the partial image block and then calculate the sum of information. Extract several images and calculate their color histogram information as the first step. Calculate the Euclidean Distance of the color histogram information between sample image and each image in the database. Then search the similar images with PSO (Particle Swarm Optimization Algorithm). If the search results are satisfactory, output the images, otherwise re-initialize the particle swarm and continue to search until you get the results that you are pleased with and at last output the images.