Automatic Image Annotation Using Bag of Words

Due to increase using images in different life application especially internet, recently many researchers interested in understanding in web and images. Automatic image annotation means attaching words, keywords or comments to an image. The inputs for image annotation system are features which are extracted from image. In this paper, a new algorithm for automatic image annotation using bag of words (BOW) and SIFT descriptor is presented. Considering the high dimensionality of SIFT features and to achieve satisfying efficiency, we apply dimension reduced technique PCA-SIFT and K-Means algorithm. Experimental results based on the images of Corel5k dataset show that the proposed method has better performance in precision and time measures.