Wavelet Transform Based Face Recognition Using SURF Descriptors
Musa M. Ameen 1,
Alaa Eleyan 2, and
Gülden Eleyan 2
1. Computer Engineering Department, Mevlana University, Konya, Turkey
2. Mechatronics Engineering Department, Mevlana University, Konya, Turkey
2. Mechatronics Engineering Department, Mevlana University, Konya, Turkey
Abstract—This paper presents a new approach to classify salient points as belonging to objects in the face classes. Speeded-Up Robust Features (SURF) algorithm is used to generate feature descriptors. SURF is applied on the subband images generated using either DWT Discrete Wavelet Transform (DWT) or Gabor Wavelet Transform (GWT). The matching scores obtained from these subband images are fused and final decision is reached. Experiments results indicate that our proposed approach showed increase in the performance for both wavelet transforms when applied on the well-known ORL and PUT face databases.
Index Terms—face recognition, speeded-up robust features, scale-invariant feature transform, discrete wavelet transform, Gabor wavelet transform
Cite: Musa M. Ameen, Alaa Eleyan, and Gülden Eleyan, "Wavelet Transform Based Face Recognition Using SURF Descriptors," International Journal of Electronics and Electrical Engineering, Vol. 5, No. 1, pp. 94-98, February 2017. doi: 10.18178/ijeee.5.1.94-98
Cite: Musa M. Ameen, Alaa Eleyan, and Gülden Eleyan, "Wavelet Transform Based Face Recognition Using SURF Descriptors," International Journal of Electronics and Electrical Engineering, Vol. 5, No. 1, pp. 94-98, February 2017. doi: 10.18178/ijeee.5.1.94-98
Array
Previous paper：Smart Phone Based Energy Monitoring System for 3 Phase Induction Motors
Next paper：Last page
Next paper：Last page