An Effective Clustering Using Moth Flame Optimization for Content-Based Image Retrieval

Content-based image retrieval plays an important role in many domains and the volume of the image database increasing tremendously. It is very complex to comparing the query image feature with all images in the dataset during the retrieval phase. Due to computational complexity increase which degrades the performance of recognition accuracy. The proposed model Clustering using Moth Flame Optimizer are pre-processing, feature extraction, feature fusion, clustering and classification. The corel lk dataset is utilized in this research for effective classification of CBIR method. Filter using First Average Peer Group (F APG) is used to remove the noise in the pre-processing stage. Feature extracted using Local Binary Pattern (LBP) and Hue Saturation Value (HSV) based feature extraction techniques. All these two features are fused into single feature using average and weighted technique. Next, proposed model clustering using Moth Flame Optimization algorithm, it enhance the flames and flame value is obtained in MFO.Lastly, the relevant image is retrieved using SAR classifier which utilized Deep Learning (DL) for effective classification. The obtained result shows the proposed clustering using MFO model achieve better accuracy of 99.87% on Corel lk dataset which insure accurate classification compared to other existing methods like Bi-Iayer method and Texture from semantic pyramid.