A novel approach using T.H.E.S methodology for CBIR

A search engine is expected to extract results with the high end accuracy .The basic motto behind the image search engines are no different. It should retrieve the best possible match from the database for the query made. But, the accessible search engines are based on the keywords only. In order to...

Full description

Saved in:
Bibliographic Details
Main Authors: Shriram, K. V., Priyadarsini, P. L. K., Gokul, M., Subashri, V., Sivaraman, R.
Format: Conference Proceeding
Language:English
Subjects:
CBIR
descriptor
detector
Entropy
GLCM
histogram
Histograms
Image color analysis
Image search
interest points
Region of interest
SURF
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A search engine is expected to extract results with the high end accuracy .The basic motto behind the image search engines are no different. It should retrieve the best possible match from the database for the query made. But, the accessible search engines are based on the keywords only. In order to overcome the inefficiencies in this approach, we are proposing a search based on CBIR (content based image retrieval). This paper discusses, in depth, about the methodologies to construct an efficient CBIR and also presents a novel approach which performs with increased efficiency in many aspects. We have used OpenCV, an image processing toolbox for implementing and testing our concepts.
DOI:10.1109/ICSIPR.2013.6497781