Gabor-HOG Features based Face Recognition Scheme

Hamid Ouanan (Moulay Ismail University)
Mohammed Ouanan (Moulay Ismail University)
Brahim Aksasse (Moulay Ismail University)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 15, No 2: August 2015

Extraction of invariant features is the core of Face RecognitionSystems (FRS). This work proposes a novel feature extractor-fusion scheme using two powerful feature descriptor known as Gabor Filters (GFs) and Histogram of Oriented Gradient (HOG), which the face image is filtered with the multiscale multiresolution Gabor filter bank to generate multiple Gabor magnitude images (GMIs), then the down-sampled GMIs and apply Histogram of Oriented Gradient to form the features. The experimental results on the FERET face database show the effectiveness of our methods.