Content Based Image Retrieval Using Shape and Color Feature.

Date of Submission

December 2002

Date of Award

Winter 12-12-2003

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science


Electronics and Communication Sciences Unit (ECSU-Kolkata)


Chanda, Bhabatosh (ECSU-Kolkata; ISI)

Abstract (Summary of the Work)

The term "Content Based Images Retrieval System" has two important parts: Content of images and Image retrieval.Content of image:An image can be describe by several features, which include shape, color, texture and spatial relationship are called content of image. By using these features, images can be modeled as well as can be indexed in an image database. An image retrieval system which is based on these low level features are called Content Based Images Retrieval (CBIR) System.Image Retrieval: How to retrieve images similar to a query-image is becoming an important problem with the rapid increase of media information on the web. Generally, users wants to provide query images and use some kind of system which can present their with a set of similar images. Therefore, how to describe and model an image, how to compare different images and judge whether they are similar, how to construct an index of image database, and how to conduct search efficiently are some of the key problems in any so-called image retrieval systemInterest in the potential of digital images has increased enormously over the last few years, fuelled at least in part by the rapid growth of imaging on the World-Wide Web. Users need to retrieve: images from a collection come from a variety of domains, including crime prevention (e.g. finger print identification, face recognition), medicine (e.g. case-based diagnosis from radiographs or scan data), education, architecture, fashion, publishing, and so on. This images are stored may be in a database or may be on many databases scattered over various geographical locations. So a huge information is out there and can easily be accessed. Information can not be accessed unless it is properly organized, because searching and locating a desired piece of image from a varied and large collection usually result in a total frustration [1). Two major research communities, Database Management and Computer Vision, are putting considerable effort towards the solution of this problem. Accordingly two major approaches have emerged: text based and visual based image retrieval.In text-based Database Management System, images are first manually annotated using a sat of keywords that describe the content of the image. Actually here images are indexed and arranged using these keywords, finally images are retrieved based on a text-based query. This text-based image retrieval techniques face two major problem: a labour intensiveness and annotation impreciseness.(CBIR operates on a totally different principle from keyword indexing. Primitive features characterizing image content, such as color, texture, and shape, are computed for both stored and query images, and used to identify (say) the 20 stored images that are most closely matching the query.1.1 What is CBIR system?It would appear that Kota was the first person to use the term content based image retrieval. He used it describe his experiments about automatic retrieval of images from large image database, by color and shape features. Since then, the term has been widely used to represent the process of retrieving desired images on the basis of the features (such as color, shape and texture) which can be automatically extracted from the images themselves.CBIR differs from classical information retrieval in that image database are essentially unstructured. One key issue with any kind of image processing is the need to extract useful information from the raw data. Image databases thus differ fundamentally from text databases, where the raw material (word stored as ASCII character string) has already been logically structured by the another. There are basic issues to be consider with CBIR system:• Understanding image users' needs and information seeking behavior.• Identification of suitable ways of describing image content.• Extracting features from raw images.• Providing compact storage for large image database.• Matching query and stored images in a way that reflects human similarity judgtmetits.Efficiently accessing stored images by content.Providing usable human interface to CBIR system.1.2 InterfacingA tool that helps the users to express their search needs accurately and easily is crucial in any retrieval system. Images retrieval system is no exception of this. Some system e.g. GRIM-DBMS (2] use SQL-like query language.


ProQuest Collection ID:

Control Number


Creative Commons License

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


This document is currently not available here.