Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/11511
Authors: Gali, Raghupathi
Issue Date: 2010
Abstract: Content-based image retrieval (CBIR) techniques, based on the low-level image content features, enable a powerful approach in retrieving images. Most of the CBIR techniques extract significant features from the images to construct image-related feature vectors and then store the feature vectors in database. Thus, the search for target images depends upon the comparison of the feature vectors between the query and the ones in database. A similarity, measurement is performed to determine how similar the images in database to the query are in terms of their visual contents. Besides, these features with the help of query-by-example (QBE), the users can search for the target images by providing an individual example. In the present work for CBIR system, all the image feature descriptors including color descriptors, texture descriptors and shape descriptors are used to represent low-level image features. Implementation of one feature descriptor doesn't give sufficient retrieval accuracy. For combining of different types of features, there is a need to train these features with different weights to achieve good results is sufficient. A real coded chromosome genetic algorithm (GA) and any performance evaluation parameter of CBIR like precision or recall is used as fitness function to optimize feature weights. Meanwhile, a real coded chromosome corresponding to higher precision as fitness function is used to select optimum weights of features.
Other Identifiers: M.Tech
Research Supervisor/ Guide: Anand, R. S.
Dewal, M. L.
metadata.dc.type: M.Tech Dessertation
Appears in Collections:MASTERS' THESES (Electrical Engg)

Files in This Item:
File Description SizeFormat 
EEDG20667.pdf3.78 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.