Lexical Tag Parsing, Contour Filter Refine and Multilevel Annotation Techniques for Extracting Relevant Cartoon Images
C. Menaka M. C. A1, N. Nagadeepa2

1C. Menaka M.C.A, M.Phil. Research Scholar, Bharathiar University, Coimbatore (Tamil Nadu), India.
2Dr. N. Nagadeepa, Principal, Karur Velalar College of Arts and Science Women, Karur (Tamil Nadu), India.
Manuscript received on 14 August 2017 | Revised Manuscript received on 20 August 2017 | Manuscript Published on 30 August 2017 | PP: 13-17 | Volume-6 Issue-12, August 2017 | Retrieval Number: L24550861217/17©BEIESP
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 (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Many number of techniques are used in the existing systems to classify the images in the process of web image classification. In this work, proposed technique considers two HTML tags namely alt and src. In a group of web pages these tags are taken into account to download the images. Mainly this approach considers the cartoon image category web link then images can be extracted and stored. LTP techniques is applied here to parse the given tags. Images are clustered and stored in their respective folders as per the category after clustering process. CFR algorithm is used here to refine the images for storing. MIA technique is applied here to give annotation for all images which is in the cluster for best retrieval. Finally based upon the given input as image resultant image can be searched from various available clusters and return to the user along with its detailed description.
Keywords: Image Clustering, LTP, MIA, CFR, Image annotation, SIC.

Scope of the Article: Image Security