Gender Classification Using Appearance Based Face Recognition Methods

Abstract

Today, intensive studies on facial recognition systems have become an important issue. Gender classification, which is one of the basic approaches of human computer interaction, is widely used in many areas, from smart building applications to security investigations. In this study, gender recognition has been made by using appearance based gender classification systems. Local Binary Pattern (LBP), Radon and Gabor transform methods have been used to extract of features in appearance based systems. Principal Component Analysis (PCA) method has been preferred to reduce the high dimension of the resulting data matrices. Support Vector Machine (SVM) classifier is used to classify data. The front face views of the people in the FERET database were used as a database. Accuracy rates of up to 89% were achieved when 70% of the images in the database were used as training data. This value reaches up to 96% were obtained when 90% of the images were used as training data. Additionaly, the results was showed that the inclusion of Radon conversion in current gender recognition methods increases system accuracy.

Keywords

• Gender recognition
• Appearance based methods
• LBP
• PCA

Görünüm Tabanlı Yüz Tanıma Yöntemleri Kullanılarak Cinsiyet Belirleme

Öz

Teknolojik gelişmeler ile birlikte yüz ve cinsiyet tanıma sistemleri günümüzün popüler çalışmalar konusu haline gelmiştir. İnsan bilgisayar etkileşiminin temel yaklaşımlarından biri olan cinsiyet sınıflandırması, akıllı bina uygulamalarından güvenlik soruşturmalarına kadar pek çok alanda yaygın olarak kullanılmaktadır. Bu çalışmada, görünüm tabanlı cinsiyet sınıflandırma yöntemleri kullanılarak cinsiyet tespiti yapılmıştır. Görünüm tabanlı sistemlerde özellik çıkarmak için yerel ikili örüntü (LBP), Radon ve Gabor dönüşümleri kullanılmıştır. Ortaya çıkan veri matrislerindeki yüksek boyutları azaltmak için ise temel bileşen analizi (PCA) yöntemi tercih edilmiştir. Verileri sınıflandırmak için destek vektör makinesi (SVM) sınıflandırıcısı kullanılmıştır. Veri tabanı olarak FERET veri tabanındaki kişilere ait ön yüz görünümleri kullanılmıştır. Veri tabanındaki resimlerin %70’i eğitim verisi olarak kullanıldığında %89; %90’ı eğitim verisi olarak kullanıldığında ise %96’lara varan doğruluk oranlarına ulaşılmıştır. Ayıca sonuçlardan Radon dönüşümünün mevcut cinsiyet belirleme yöntemlerine dahil edilmesinin sistem doğruluğunu artırdığı görülmüştür.

Anahtar Kelimeler

• Cinsiyet tanıma
• Görünüm tabanlı yöntemler
• LBP
• PCA
• SVM

Kaynakça

1. Ahonen, T., Hadid, A. Pietikinen, M. (2004). Face Recognition with Local Binary Patterns. The 8th European Conference on Computer Vision, 11-14 May, Prague, Czech Republic.
2. Alexandre, L.A. (2010). Gender recognition: A multiscale decision fusion approach. Pattern Recognition Letters, Cilt 31, Sayı 11, ss. 1422-1427.
3. Azarmehr, R., Laganiere, R., Lee, W.S., Xu, C., Laroche, D. (2015). Real-time embedded age and gender classification in unconstrained video. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 7-12 June, Boston, USA.
4. Dhanashri P.L., Kailash J.K. (2016). Gender Classification using Face Image: A review. International Journal of Latest Trends in Engineering and Technology, Cilt 7, Sayı 2, ss. 333-337.
5. Fang, Y., Wang, Z. (2008). Improving LBP features for gender classification. International Conference on Wavelet Analysis and Pattern Recognition, 30-31 August, Hong Kong, China.
6. Jain, A., Huang, J. (2004). Integrating independent components and linear discriminant analysis for gender classification. Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 19 May, Seoul, South Korea.
7. Iga, R., Izumi, K., Hayashi, H., Fukano, G., Ohtani, T. (2003). A gender and age estimation system from face images. SICE 2003 Annual Conference, 4-6 August, Fukui, Japan.
8. Kalam, S., Guttikonda, G. (2014). Gender classification using geometric facial features. International Journal of Computer Applications, Cilt 85, Sayı 7, ss. 32-37.

9. Khalifa, T.A.M. (2016). Predicting Age and Gender of People by Using Image Processing Techniques. Computer Engineering Atilim University, Master Thesis.
10. Lian, H.C., Lu, B.L. (2006). Multi-view gender classification using local binary patterns and support vector machines. The third International Conference on Advances in Neural Networks, 28 May 28 – 1 June, Chengdu, China.
11. Liu, H., Gao, Y., Wang, C. (2014). Gender identification in unconstrained scenarios using Self-Similarity of Gradients features. IEEE International Conference on Image Processing, 27-30 October, Paris, France.
12. Lu, L., Shi, P. (2009). A novel fusion-based method for expression-invariant gender classification. IEEE International Conference on Acoustics, Speech and Signal Processing, 19-24 April, Taipei, Taiwan.
13. Mäkinen, E., Raisamo, R. (2008). An experimental comparison of gender classification methods. Pattern Recognition Letters, Cilt 29, Sayı 10, ss. 1544–1556.
14. Moghaddam, B., Yang, M.H. (2002). Learning Gender with Support Faces. IEEE Transactions on Pattern Analysis and Machine Intelligene, Cilt 24, Sayı 5, ss. 707-711.
15. Ojala, T., Pietikäinen, M., Mäenpää, T. (2002). Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, Cilt 24, Sayı 7, ss. 971–987.
16. Öztürk, E., Kurnaz, Ç. (2018). Gender Recognition System from Face Images with Artificial Neural Networks. International Eurasian Conference on Science Engineering and Technology. 22-23 November, Ankara, Turkey.
17. Ramesha, K., Raja, K.B., Venugopal, K.R., Patnaik, L.M. (2010). Feature Extraction-Based Face Recognition, Gender and Age Classification. International Journal of Computer Theory and Engineering, Cilt 2, Sayı 5, ss. 798-820.
18. Uzun, M., Gökmen, M. (2018). Face Recognition with Local Walsh Transform. Signal Processing: Image Communication, Cilt 61, ss. 85-96.
19. Wang, C., Huang, D., Wang, Y., Zhang, G. (2012). Facial Image-Based Gender Classification Using Local Circular Patterns. The 21st International Conference on Pattern Recognition, 11-15 November, Tsukuba, Japan.
20. Xie, S., Shan, S., Chen, X., Chen, J. (2010). Fusing Local Patterns of Gabor Magnitude and Phase for Face Recognition. IEEE Transactions on Image Processing, Cilt 19, Sayı 5, ss. 1349-1361.
21. Yang, Z., Ai, H. (2007). Demographic classification with local binary patterns. The International Conference on Advances in Biometrics, 27-29 August, Seoul, Korea.
22. Zafeiriou, S., Tefas, A., Ioannis Pitas, I. (2008). Gender Determination Using a Support Vector Machine Variant. 16th European Signal Processing Conference, 25-29 August, Lausanne, Switzerland.
23. Zhang, J., Tan, T., Ma, L. (2002). Invariant Texture Segmentation via Circular Gabor Filters. The 16th IAPR International Conference on Pattern Recognition, August, Quebec, Canada.
24. Zhang, W., Shan, S., Gao, W., Chen, X., Zhang, H. (2005). Local Gabor Binary Pattern Histogram Sequence (LGBPHS): A Novel Non-Statistical Model for Face Representation and Recognition. The tenth IEEE International Conference on Computer Vision. 17-21 October, Beijing, China.
25. Zhang, B., Shan, S., Chen, X., Gao, W. (2007). Histogram of Gabor Phase Patterns (HGPP): A Novel Object Representation Approach for Face Recognition. IEEE Transactions on Image Processing, Cilt 16, Sayı 1, ss. 57–68.