An Optimized CBIR Using Particle Swarm Optimization Algorithm
subhakala subbaiyan · 2017
Storage and retrieval of images over a large database is an important issue. Content Based Image Retrieval system provides solution for this issue. In Content Based Image Retrieval(CBIR) similar images are retrieved using low level features such as color, texture, edge, etc that are extracted both from the query image and the database. In CBIR less amount of retrieval time with high accuracy is desired property. The proposed system achieves this property by using Particle Swarm Optimization algorithm. The proposed system consists of the following phases (i) Color feature extraction using (luminance(y), blue chrominance (u), red chrominance (v)) method (ii) Texture feature extraction using Grey Level Co-occurrence Matrix (iii) Edge feature extraction using Edge Histogram Descriptor (iv) Measurement of Similarity between Query image and the Database image using Euclidean Distance. (v) Optimization of retrieved result using Particle Swarm Optimization. In comparison with the existing approach, the proposed approach significantly improves the precision and recall of the retrieval system.