Image Search Engine for Retrieval of Similar Images Using CBIR, SVM, SIFT

摘要

Feature selection and extraction is an important aspect of any image based application. There are lots of methods that are used for selecting and extracting features from the image. Features are nothing but the keypoints which gives us brief information about any image. Selection of these methods depends upon the research under consideration e.g. medical image needs different features extraction methods than general purpose image application. There are many methods of CBIR based on low level and high level features. As compare to low level feature high level features i.e. frequency domain are giving best result because they are extracting more details about the image.