Studies on Content Based Image Retrieval System (CBIR) with Relevance Feedback Using Neural Networks and MPEG-7 Features.

Date of Submission

December 2009

Date of Award

Winter 12-12-2010

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science

Department

Machine Intelligence Unit (MIU-Kolkata)

Supervisor

Kundu, Malay Kumar (MIU-Kolkata; ISI)

Abstract (Summary of the Work)

Effective image retrieval from a large database is a difficult problem and is still far from being solved. Hence, the retrieval of relevant images, based on measuring the similarity between automatically derived features(color, texture and shape, etc) of the query image and that of the images stored in the database, aproblem popularly known as content based image retrieval is a highly challenging task.. In a conventional CBIR approches, an image is usually represented bya set of features, where the feature vector is a point in a multidimensional space. It also has numerous applications in areas like Biomedicine (X ray, Pathology, CT, MRI,..), Crime Management, Commercial (Fashion, Catalogue, Design, Journalism,..) and it is widely used in medical services. Although extensive research has been performed on the CBIR over several decades, we have yet to acheive the accuracy from a fully automated CBIR system. In this approach we have used MPEG 7 features to extract the image information with relevance feedback using neural networks. The algorithm implemented in c.

Comments

ProQuest Collection ID: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:28843195

Control Number

ISI-DISS-2009-228

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

DOI

http://dspace.isical.ac.in:8080/jspui/handle/10263/6386

This document is currently not available here.

Share

COinS