Content Based Image Retrieval using Feature Extraction with Machine Learning

Content-Based image Retrieval (CBIR) is a technique of image retrieval which uses the visual features of an image such as color, shape and texture in order to search the user based query images from the large databases. CBIR depends on feature extraction of an image which are the visual features and these features are extracted automatically i,e without human interaction. In this paper SIFT feature extraction algorithm is used for feature extraction, which basically gives us the key point in an image. SIFT image feature algorithm give a set of image features that are not valuable so we use the optimization technique BFOA (Bacteria foraging optimization algorithm) to reduce the complexity, cost, energy and Time consumptions. Then for similarity check a deep neural network is trained and then the validation and texting phases are done accordingly which lead to a better performance as compared to previously done techniques. The accuracy rates are relatively excellent in the proposed technique.