Content based Image Search Engine with Grey Wolf Optimized Neural Network

The systems with Content-based image retrieval (CBIR) approach enable the search and retrieval of images that are comparable to a particular query image by utilizing attributes that indicate the content visualization of such images. The present study created a mechanism for recovering images indexed in databases based on their visual information without the necessity for linguistic annotations. Feature vectors were obtained from the hybrid of DWT and LBP visual contents. The classification is achieved by using Grey Wolf optimized Neural Network (GWO-NN) technique. The proposed strategy yielded 95.59% accurate findings on publicly available databases.