Design and development of systems for image segmentation and content based image retrieval

Color, texture and shape information have been the primitive image descriptors in content based image retrieval systems. Here novel framework for combining all the three i.e. color, texture and shape information, to achieve higher retrieval efficiency is presented. The image is partitioned into non overlapping tiles of equal size. The color moments and geometric moments serve as local descriptors of color and texture respectively. This local information is captured for two resolutions and two grid layouts that provide different details of the same image. Shape information is captured in terms of edge images computed using Image Segmentation for shape matching. The combination of the color, texture and shape features provide a robust feature set for image retrieval. The results are compared with standard methods, like histogram based and SIMPLICITY, on the basis of precision and time required for retrieval.