An integrated filter based approach for image abstraction and stylization

Nagendra Swamy, H. S. and Pavan Kumar, M. P. (2013) An integrated filter based approach for image abstraction and stylization. In: Multimedia Processing, Communication and Computing Applications. Springer, New Delhi, pp. 241-252. ISBN 978-81-322-1143-3

Full text not available from this repository. (Request a copy)
Official URL: http://doi.org/10.1007/978-81-322-1143-3_20

Abstract

In this paper, we present a non-photo-realistic image rendering (NPR) technique based on integrated filtering approach. The proposed method integrates 2D anisotropic filter, 2D difference of Gaussian filter, modified coherence shock filter and mean curvature flow (MCF). Coherence shock filter is applied iteratively to enhance the edge information in an image. Dithering with a fixed deviation value is also applied to produce a rendering effect in the abstracted image. The proposed method can be applied to color as well as gray scale images to produce stylized and cartoon like images. The method does not require any kind of post processing for image abstraction. Implementation of the proposed work is carried out in Mat Lab environment using local library functions. Efficacy of the proposed work has been corroborated by conducting experiments on various types of images and the results have also been compared with the other contemporary work. The approach is found to be computationally efficient in producing effective cartoon like images being simple in terms of its implementation.

Item Type: Book Section
Uncontrolled Keywords: Anisotropic filters, Anisotropy, Communication, Dithering, Iterative methods, Mean curvature flow, Non-Photorealistic Rendering, Object recognition, Shock filters, Signal filtering and prediction
Subjects: D Physical Science > Computer Science
Divisions: Department of > Computer Science
Depositing User: Arshiya Kousar
Date Deposited: 21 Oct 2019 06:14
Last Modified: 21 Oct 2019 06:14
URI: http://eprints.uni-mysore.ac.in/id/eprint/9143

Actions (login required)

View Item View Item