SOM based Face Recognition using Steganography and DWT Compression Techniques

Downloads

Downloads per month over past year

Sujatha, B.M. and Nayina Ramapur, . and Shubhangi Lagali, . and Suresh Babu, K and Raja, K.B. and Venugopal, K.R. (2016) SOM based Face Recognition using Steganography and DWT Compression Techniques. International Journal of Computer Science and Information Security, 14 (9). p. 806. ISSN 1947-5500

[img]
Preview
Text
86_Paper_310816157_IJCSIS_Camera_Ready_pp._806-826-with-cover-page-v2.pdf

Download (2MB) | Preview

Abstract

Biometrics is used in day to day life of human beings to access mobile phones, computers, vehicles etc. In this paper, we propose SOM based Face Recognition using Steganography and DWT Compression Techniques. The various available standard face databases such as ORL, JAFFE, NRI, YALE, Indian male and Indian female are used to test the performance of the algorithm. The number of face images per person is reduced using stenography compression technique. The preprocessing techniques such as resize and Gaussian filter are applied on reduced number of face images for uniform size and good quality of images. The initial features are extracted using Discrete Wavelet Transform (DWT) and Local Binary Pattern (LBP). The DWT and LBP features are fused using arithmetic additions and connected to input of Self Organizing Map (SOM). The final features are extracted from SOM. The test image features are compared with face database images using Euclidian Distance (ED) to compute performance parameters. It is observed that the performance of the proposed method is better than the existing methods.

Item Type: Article
Subjects: Faculty of Engineering > Computer Science & Information Science Engineering
Divisions: University Visvesvarayya College of Engineering > Department of Computer Science and Information Science Engineering
Depositing User: Ms. Shwetha A C
Date Deposited: 30 Sep 2021 09:15
Last Modified: 30 Sep 2021 09:15
URI: http://eprints-bangaloreuniversity.in/id/eprint/9760

Actions (login required)

View Item View Item