Authors : Rucha S. Patil and Avinash J. Agrawal
Abstract: An image retrieval system is one of the important computer systems for browsing and retrieving images from a large database. There are two approaches for image retrieval, Text Based Image Retrieval (TBIR) and Content Based Image Retrieval (CBIR). The drawback of TBIR is manual annotation which is impossible and expensive task for large database. The problems in TBIR have raised the interest of researchers to come up with techniques for retrieving images. Image annotation task assign a set of semantic tags or keywords to an image based on some models learned from certain training data. Automatically assigning keywords to images is of great interest as it is an active topic of research in computer vision and pattern recognition. In this study, we proposed a framework for content based image retrieval using k-means clustering and also performed image annotation using adaptive threshold.
Rucha S. Patil and Avinash J. Agrawal, 2017. Image Retrieval using K-Means Clustering and Image Annotation. Journal of Engineering and Applied Sciences, 12: 8021-8024.