"Sketch Based Image retrieval from large database using geometric approach"

Authors

G.G. Rajput, Akkamahadevi Women’s University, Vijayapura – 586108, Karnataka, India.

Prashantha, Rani Channamma University, Belagavi – 591156, Karnataka India.

Abstract

Sketch based image retrieval is an efficient method for the retrieval of images from the database matching the sketch query. Drawing a sketch does not require the user to have a high skill and that it is a reliable approach compared to text-based image retrieval. Sketch-to-image matching on a large-scale image database should ensure the retrieval of related images for a given sketch query. Efficient sketch-based image retrieval from large database is presented in this paper. Features are extracted by slicing the contour of the image using angle and radial partitioning approach. Different distance measures are used to measure the distance between sketch and database feature. The experiment is evaluated on flickr15k dataset. The proposed method shows effective retrieval performance compared to other methods present in the literature.