Clustering of Digital Images using Shape Features with SOM
Mayank Singh1, Viranjay M. Srivastava2

1Mayank Singh, Department of Electrical Electronic & Computer Engineering, Howard College, University of KwaZulu Natal, Durban South Africa. 

2Viranjay M. Srivastava, Research Scholar, BITS, Pilani, India. 

Manuscript received on 10 October 2019 | Revised Manuscript received on 24 October 2019 | Manuscript Published on 26 December 2019 | PP: 918-922 | Volume-8 Issue-12S October 2019 | Retrieval Number: L120210812S19/2019©BEIESP | DOI: 10.35940/ijitee.L1202.10812S19

Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (

Abstract: In these days people are interested in using digital images. So the size of image databases is increasing rapidly. It leads retrieval problem of images from large databases. Machine learning algorithms are applying in recent research to simplify the task of image retrieval and make it automatic. Thus the concept of content based image retrieval system came into existence. In this system the images are extracted based on similar content. Content means features of the images and it is formed by feature extraction of the images in databases. Contents can be edges, color, shape, gradient, orientation, histogram gradient etc. These contents are clustered into various groups of similar feature vectors. So for any input image the selected feature is searched for and image is retrieved from the database. This reduces the time complexity. There have been many algorithms for implementing the content based image retrieval system. In this research work we propose a novel paradigm where in shape features are extracted from the database images and are used to train the self-organizing map to cluster the shape features. These clusters are then used for image retrieval.

Keywords: Digital Image Clustering; Clustering Techniques; Content based Image Retrieval; Shape Features; Self-Organizing Map.
Scope of the Article: Clustering