Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi
Yazarlar: Tariq KHALIFA, Gökhan ŞENGÜL
Konular:Bilgisayar Bilimleri, Bilgi Sistemleri
DOI:10.28948/ngumuh.383746
Anahtar Kelimeler:Facial images,Gender prediction,Local Binary Patterns,Histograms of Oriented Gradients
Özet: Gender prediction from facial images can be used in a large number of applications including human-computer interaction, customer information measurement, access control, etc. Furthermore, it can substantially effect on many fields, such as security systems, biometric authentication, medical imaging systems, demographic studies, content based searching, and surveillance system. In this study, we proposed to use Local Binary Patterns (LBP) and Histograms of Oriented Gradients (HOG) as the feature extractor and k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) as the classifier in order to predict the gender of the people from facial images. We tested the proposed method in FERET and UTD databases. We used leave-one-out approach as the cross validation technique. The results are promising.