Content Based Image Retrieval System - Using Color, Shape & Texture

Abstract

Rapid increase in the usage of color image has led to the need of an image retrieval system for them. The world is dominated by visual information and a tremendous amount of such information is being added every day. It would be impossible to cope with this explosion of visual data, unless they are organized in such a manner that we can retrieve them efficiently and effectively. The main problem in organizing and managing such visual data is indexing, the assignment of a synthetic descriptor which facilitates its retrieval. There are various mathematical models that describe the image information and content. The Work presents the image retrieval based on color, shape and texture. Two different indexed data bases are maintained for the mentioned features. Global color dominance in the image is used as the index key in the color database, and local color information is retained by quantizing images in color and pixel space to optimize in memory need and performance. Texture information is obtained using co-occurrence matrix of the color images converted to gray. The properties like correlation, energy, homogeneity and contrast are good parameters to measure textural property; these are evaluated at proper offset and angle. Shape in the image is identified using the contours in the image. Only the geometrical figures like circle, ellipse, quadrilateral, triangle, pentagon etc. can be identified. While retrieving images based on features like texture and shape one thing missing is, color. Color is inevitable feature that dominates the human perception most. To a human observer, images with similar color scheme and reasonably similar in other features is voted higher in similarity list, than images measuring better in other feature descriptors. Making this as the base, the work presented here uses all the three features independently and also grouping color with texture and shape. Results have shown that they match better with the human perception. MATLAB 7 Is the tool used to implement image processing algorithms and postgre SQL 8.1 for windows is the DBMS used for managing the database of image features.