A Novel Approach to Enhance Face Feature Extraction using Pencil Sketches
Anil J.1, L. Padma Suresh2

1Anil J., Department of Electrical & Electronics Engineering, Noorul Islam University, Noorul Islam College for Higher Education, Thuckalay (Tamil Nadu), India.
2Dr. L. Padma Suresh, Department of Electrical & Electronics Engineering, APJ Abdul Kalam Technological University, Baselios Mathews II College of Engineering, Kollam (Kerala), India.
Manuscript received on 26 May 2019 | Revised Manuscript received on 13 June 2019 | Manuscript Published on 26 June 2019 | PP: 303-311 | Volume-8 Issue-1S5 June 2019 | Retrieval Number: A00530681S519/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Feature extraction is the primary step in all the recognition problems. For recognizing a face, the facial features should be extracted. The recognition rate depends on the accuracy in feature extraction. In this paper a method for enhancing face feature extraction has been proposed. The method proposed has implemented a Double Sized Scaled Horizontal Vertical Oblique (DSHVO) filter before extracting the feature points from the face. The application of DSHVO filter converts the input image to an image resembling a pencil sketch. In this paper different feature extraction methods like SURF, Harris and FAST are carried out in gray scale images and the results obtained are compared. The feature extraction is also carried out after the application of the DSHVO filter by using these feature extraction methods. Experimental results show that the conversion of gray scale images to pencil sketch images have considerably improved the feature extraction accuracy. The performance of these feature extraction methods in gray scale and pencil sketch images is done and the results are compared. Based on the results obtained the feature extraction methods are evaluated.
Keywords: FAST, Feature Extraction, Gray Scale, Harris, Pencil Sketch, SURF.
Scope of the Article: Reflection and Metadata Approaches