A New SIFT-Based Image Descriptor Applicable for Content Based Image Retrieval
Davar Giveki, Mohamad Ali Soltanshahi, Fatemeh Shiri, Hadis Tarrah, Received January · 2015
The large amounts of image collections available from a variety of sources have posed increasing technical challenges to computer systems to store/transmit and index/manage the image data to make such collections easily accessible. To search and retrieve the expected images from the data base a Content Based Image Retrieval (CBIR) system is highly demanded. CBIR extracts features of a query image and try to match them with extracted features from images in the data base. This paper introduces two novel methods that can be used as image descriptors. The basis of the proposed methods is built upon Scale Invariant Feature Transform (SIFT) method. After extracting image features using SIFT, k-means clustering is applied on feature matrix extracted by SIFT, and then two new kinds of dimensionality reduction reapplied to make SIFT features more efficient and realistic for image retrieval problem. Using the proposed strategies we can not only take the advantage of SIFT features but we can also highly decrease the memory storage used by SIFT features. Finally, proposed methods are compared with some other state of the art methods and as a result, our proposed retrieval system is faster and more accurate. Experimental results on two popular datasets, Corel (includes 1000 images) and OT data sets (includes 2688 images), show the superiority and efficiency of the proposed methods.