Content Based Image Retrieval using Color and Shape Features

Abstract: This paper proposes, a method is used for Content-Based Image Retrieval (CBIR) using color and shape features. Active research in CBIR is geared towards the development of methodologies for analyzing, interpreting, cataloging and indexing image databases. In addition to their development, efforts are also being made to evaluate the performance of image retrieval systems. The quality of response is heavily dependent on the choice of the method used to generate feature vectors and similarity measure for comparison of features. The speed of shape based retrieval can be enhanced by considering approximate shape rather than the exact shape. In addition to this a combination of color and shape based is also included to improve the accuracy of the result. We have implemented the CBIR system which takes into consideration the low level features of image which is more comprehensive when compared to high level features. Here the distance between database images and query image in found and the images with minimum distance will be displayed as output. For Shape Based Image Retrieval, the retrieved images are displayed based on the similarity of the shape of object in image with the shape of query image shape. The images are ranked depending on the similarity using Euclidean Distance i.e. more similar images are displayed at the top and the images with less similarity will be displayed later. From the comparison of Color Based Image Retrieval and Shape Based Image Retrieval result, we notice that the accuracy of Shape Based Image Retrieval is more than the accuracy of Color Based Image Retrieval. Though the complexity of Shape Based Image Retrieval is more than the Color Based Image Retrieval, Shape Based Image Retrieval consumes less time and gives more accuracy in retrieved results.

Keywords: CBIR, Shape Based Image Retrieval, Color Based Image Retrieval, Euclidean Distance.