An Efficient Image Classification using Bag-of-Words based on SURF and Texture Features

Image Classification is an important and challenging task in various applications like: video surveillance, image and video retrieval etc. It is the task of correctly assigning an image to the class to which it belongs to. Various approaches SIFT(Scale-Invariant Feature Transform), SURF, Local features have been used in literature for this purpose. SURF gives good accuracy for classification of image only if the query image is from the training set. The proposed approach in this paper is using the combination of both SURF and Texture. Even if the query image is not from training set the proposed approach gives higher accuracy of 94%.