Flag Counter
AKILLI SİSTEMLER VE UYGULAMALARI DERGİSİ
JOURNAL OF INTELLIGENT SYSTEMS WITH APPLICATIONS
J. Intell. Syst. Appl.
E-ISSN: 2667-6893
Creative Commons License This work is licensed under a Creative Commons Attribution 4.0 International License.

Using Local Features in Face Recognition Systems

Yüz Tanıma Sistemlerinde Yerel Özniteliklerin Kullanılması

How to cite: Aydın Y, Akar F. Using local features in face recognition systems. Akıllı Sistemler ve Uygulamaları Dergisi (Journal of Intelligent Systems with Applications) 2018; 1(2): 131-134. DOI: 10.54856/jiswa.201812041

Full Text: PDF, in Turkish.

Total number of downloads: 626

Title: Using Local Features in Face Recognition Systems

Abstract: Among the many applications in the field of computer vision, face recognition systems; is a subject that has been studied extensively and has been working for a long time. In general, the success of facial recognition systems, which consist of feature extraction and classifier steps, depends not only on the classifier but also on the features used. In a face recognition system, the feature selection is to obtain distinctive features for recognition of different facial images of interest. For this purpose, SIFT, SURF and SIFT + SURF features, which are unchanging features to scaling and affine transformations, are used in this study. In addition, to be able to compare with these local features, the HOG feature which is a global feature, also has been added to the study. Classification was performed using support vector machine. Experimental results show that local features are more successful than the global feature HOG.

Keywords: Local features; SIFT; SURF; SIFT+SURF; face recognition systems


Başlık: Yüz Tanıma Sistemlerinde Yerel Özniteliklerin Kullanılması

Özet: Bilgisayarla görme alanındaki birçok uygulama arasında, yüz tanıma sistemleri; üzerinde yoğun çalışmalar yapılan ve uzun zamandır çalışılan bir konudur. Genel olarak öznitelik çıkarımı ve sınıflandırıcı adımlarından oluşan yüz tanıma sistemlerinin başarısı sadece sınıflandırıcıya değil aynı zamanda kullanılan özniteliklere de bağlıdır. Yüz tanıma sisteminde öznitelik seçiminin amacı farklı yüz görüntülerinin tanınması için ayırt edici özniteliklerin elde edilmesidir. Bu amaç doğrultusunda bu çalışmada ölçeklemeye ve afin dönüşümlere karşı değişmeyen öznitelikler olan SIFT, SURF ve SIFT+SURF özniteliği kullanılmıştır. Ayrıca bu yerel özniteliklerle ile kıyaslama yapılabilmesi için global bir öznitelik olan HOG özniteliği de çalışmaya eklenmiştir. Sınıflandırma destek vektör makinesi kullanılarak gerçekleştirilmiştir. Deneysel sonuçlar yerel özniteliklerin global öznitelik olan HOG özniteliğine kıyasla daha başarılı olduğunu göstermiştir.

Anahtar kelimeler: Yerel öznitelikler; SIFT; SURF; SIFT+SURF; yüz tanıma sistemleri


Bibliography:
  • Kolap AD, Shrikhande SV, Jagtap NK. Review on various face recognition techniques. International Journal of Innovative Research in Computer and Communication Engineering 2015; 3(3): 2398-2404.
  • Liau HF, Ang LM, Seng KP. A multiview face recognition system based on eigenface method. In 2008 International Symposium on Information Technology, August 26-28, 2008, Kuala Lumpur, Malaysia, pp. 1-5.
  • Satone M, Kharate GK. Selection of eigenvectors for face recognition. International Journal of Advanced Computer Science and Applications (IJACSA) 2013; 4(3).
  • Dagher I, Nachar R. Face recognition using IPCA-ICA algorithm. IEEE Transactions On Pattern Analysis and Machine Intelligence 2006; 28(6): 996-1000.
  • Eleyan A, Demirel H. Co-occurrence based statistical approach for face recognition. In 2009 24th International Symposium on Computer and Information Sciences, September 14-16, 2009, Guzelyurt, Northern Cyprus, pp. 611-615.
  • Ghazali KH, Mansor MF, Mustafa M, Hussain A. Feature extraction technique using discrete wavelet transform for image classification. In 2007 5th Student Conference on Research and Development, December 11-12, 2007, Selangor, Malaysia, pp. 1-4.
  • Mazloum J, Jalali A, Amiryan J. A novel bidirectional neural network for face recognition. In 2012 2nd International eConference on Computer and Knowledge Engineering (ICCKE), October 18-19, 2012, Mashhad, Iran, pp. 18-23.
  • Li YA, Shen YJ, Zhang GD, Yuan T, Xiao XJ, Xu HL. An efficient 3D face recognition method using geometric features. In 2010 2nd International Workshop on Intelligent Systems and Applications, May 22-23, 2010, Wuhan, China, pp. 1-4.
  • Vinay A, Hebbar D, Shekhar VS, Balasubramanya Murthy KN, Natarajan S. Two novel detector-descriptor based approaches for face recognition using SIFT and SURF. Procedia Computer Science 2015: 70: 185-197.
  • Kashif M, Deserno TM, Haak D, Jonas S. Feature Description with SIFT, SURF, BRIEF, BRISK, or FREAK? A general question answered for bone age assessment. Computers in Biology and Medicine 2016; 68: 67–75.
  • Khalifa FA, Semary NA, El-Sayed HM, Hadhoud MM, Local detectors and descriptors for object class recognition. International Journal of Intelligent Systems and Applications 2015; 7(10): 12-18.
  • Gesto-Diaz M, Tombari F, Gonzalez-Aguilera D, Lopez-Fernandez L, Rodriguez-Gonzalvez P. Feature matching evaluation for multimodal correspondence. ISPRS Journal of Photogrammetry and Remote Sensing 2017; 129: 179–188.
  • Lowe D. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 2004; 60(2): 91–110.
  • Bay H, Ess A, Tuytelaars T, Gool LV. Speeded-up robust features (SURF). Computer Vision and Image Understanding 2008; 110: 346–359.
  • Rublee E, Rabaud V, Konolige K, Bradski G. ORB: An efficient alternative to SIFT or SURF. In 2011 IEEE International Conference on Computer Vision, November 6-13, 2011, Barcelona, Spain, pp. 2564-2571.
  • Rosten E, Drummond T. Machine learning for high-speed corner detection. Proceedings of the 9th European Conference on Computer Vision, 2006, pp. 430-443.
  • Leutenegger S, Chli M, Siegwart RY. BRISK: Binary robust invariant scalable keypoints. In 2011 International Conference on Computer Vision, November 6-13, 2011, Barcelona, Spain, pp. 2548-2555.
  • Dalal N, Triggs B. Histograms of oriented gradients for human detection. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), June 20-25, 2005, San Diego, CA, USA, pp. 886-893.
  • Mikolajczyk K, Schmid C, A performance evaluation of local descriptors. IEEE Transactions On Pattern Analysis And Machine Intelligence 2005; 27(10): 1615-1630.
  • Aydin Y. Traffic sign recognition system for imbalanced dataset. MSc Thesis, Ataturk University, Erzurum, Turkey, 2016.