Home > Published Issues > 2016 > Volume 4, No. 1, June 2016 >

Implementing Discrete Wavelet and Discrete Cosine Transform with Radial Basis Function Neural Network in Facial Image Recognition

Samuel Lukas 1, Aditya Rama Mitra 2, Ririn Ikana Desanti 3, and Dion Krisnadi 1
1. Informatics Department, Universitas Pelita Harapan, Karawaci, Indonesia
2. Computer System Department, Universitas Pelita Harapan, Karawaci, Indonesia
3. Information System Department, Universitas Pelita Harapan, Karawaci, Indonesia

Abstract—Face image recognition has been widely used and implemented in many aspects of life, such as in the field of investigation or security. However, research in this area is still rarely done. Source images in this paper are taken directly from 41 students with a total of 131 faces in JPG format, each with a dimension of 256×256. By applying Discrete Wavelet Transform and Discrete Cosine Transform, an image can be represented as a number of DCT coefficients efficiently. Recognition process is done using Radial Basis Function Neural network. The experiment results show that the best configuration for RBF is 8×41×41 with recognition rate of student faces is 100% and 98% of the sample face images are identified perfectly.

Index Terms—discrete wavelet transform, discrete cosine transform, radial basis function neural network

Cite: Samuel Lukas, Aditya Rama Mitra, Ririn Ikana Desanti, and Dion Krisnadi, "Implementing Discrete Wavelet and Discrete Cosine Transform with Radial Basis Function Neural Network in Facial Image Recognition," Journal of Image and Graphics, Vol. 4, No. 1, pp. 6-10, June 2016. doi: 10.18178/joig.4.1.6-10

PREVIOUS PAPER
Enhanced Method for Face Detection Based on Feature Color
NEXT PAPER
The Face Recognition Algorithms Based on Weighted LTP