Docstoc

SIMILARITY MEASURES FOR HISTOGRAM BASED IMAGE RETRIEVAL

Document Sample
SIMILARITY MEASURES FOR HISTOGRAM  BASED IMAGE RETRIEVAL Powered By Docstoc
					   SIMILARITY MEASURES FOR HISTOGRAM BASED IMAGE
                                        RETRIEVAL



       Content-based image retrieval (CBIR) has emerged as an important area in computer
vision, multimedia computing and databases. It makes use of lower-level (primitive) features like
color, texture, spatial layout and shape, and even higher-level (semantic) features like
annotations and user interactions to retrieve images according to different search paradigms.
The most commonly used search techniques in image collections are based either on a given
example (i.e. the user provides an image query),on association (e.g. the user browses an image
collection to choose the most suitable ones), or on a given category (i.e. finding images of a
given class)

                               Although there exists a great evidence of user need for better
image data management in various fields such as crime prevention, intrusion detection, and
medical diagnosis, there is an important mismatch between the capabilities of CBIR techniques
and the needs of the users .This problem is called semantic gap and is due to the discrepancy
between semantic similarity querying (as requested by the user) and low-level features querying
(as offered by the CBIR technology).

               Efforts to reduce the gap have been focused on semantic enrichment
(interpretation) of low-level features(by getting information related either to the application
domain, the user, or the characteristics of features), integration of other sources of information
(e.g. textual data), discovery of relevant objects using kernel-based learning techniques (e.g.
support vector machines and principal component analysis), and content-based navigation using
generic links based on text or image features .
PROBLEM STATEMENT


              The Histogram Based Image Retrieval Technique is used to retrieve similar images from the
database .

 By implementing the steps mentioned below in our project we can obtain similar images related to Query
image.

             Obtaining and compute histogram for all images in terms of levels and area of each level
              using HSV Quantization.
             Computing distance functions from histogram levels of images.



        Retrieving the top fifteen similar images from the database based on different distance measures.

SYSTEM REQUIREMENT SPECIFICATION


System Requirements:



Processor                :      Celeron or Pentium processor

Operating System         :      any Operating System that support JAVA platform

Memory                   :      256MB RAM preferred

Hard Disk                :      2GB or more

Peripherals              :      Color Display Device and input devices

Languages/Platforms      :      Java Runtime Environment, Java Development Kit
(1.3.1 or higher)

Other                    :      Any Web Browser with Applet support.
Hardware Requirements:


    1. 386 Processor or Above.

    2. Minimum of 64MB RAM..

    3. Minimum 1MB Hard Disk



Software Requirements:

    1. Operating System -- Windows Professional / Windows XP/LINUX

    2. Language          -- JAVA

    3. Java web server

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:3
posted:8/21/2012
language:
pages:3