IBMRD's Journal of Management & Research

Rajivkumar Mente ¹, B. V. Dhandra ², Gururaj Mukarambi ²

Affiliations
1 Department of Computer Science, Solapur University, Solapur, IN
2 Department of P.G. Studies and Research in Computer Science, Gulbarga University, Gulbarga, IN

Abstract

With the rapid increase of multimedia information, there is a growing importance for facilitating automatic image searching and retrieval. Generally, low level visual features of the images are used in Content Based Image Retrieval (CBIR) to segment, index and retrieval of the image from the image database. Such methods may require more computational time and inefficient indexing and retrieval performance. Shape feature is among the important feature of an image since it is reflective of the human perception. Hence shape description or representation is an important issue both in object recognition and classification. Therefore an attempt is made in this paper to focus on the shape descriptor-eccentricity and color features for achieving efficient and effective retrieval performance by using kNN classifier. Experiments are carried out on proposed algorithm with 2732 images and achieved an accuracy of 98.52%.

Keywords

CBIR, Shape Descriptor, Eccentricity, KNN, RGB, Canny Edge Detection

Full Text