Çoklu Evrişimli Sinir Ağı Modelleri İle Bir Avuç İçi Damar Tanıma Yaklaşımı

Year 2021, , 237 - 242, 01.12.2021

Felix Olanrewaju Babalola , Önsen Toygar , Yiltan Bitirim

https://doi.org/10.31590/ejosat.1016532

Abstract

Bu makalede bir avuç içi damar tanıma sistemi önerilmiştir. Üç Evrişimli Sinir Ağı (CNN) modelinin (VGG16, VGG19 ve AlexNet) avuç içi damar biyometrisindeki etkisi karşılaştırılmış ve daha sonra bu modellerin Karar-Seviyesi Kaynaşımı kullanarak birleştirilmesi önerilmiştir. Bu modellerin eğitiminde çok fazla süzgeç kullanıldığı için hesaplama süresi çok yüksektir. Dolayısıyla, bu çalışmada, modellerin verimliliğini muhafaza ederken hesaplama süresini de büyük ölçüde azaltmak için modellerde kullanılan süzgeçler azaltılmıştır. Önerilen yöntem, FYO, PUT ve VERA isimli üç halka açık veritabanı kullanılarak test edilmiştir. Bireysel modellerle kıyaslandığında, önerilen yöntemin doğruluğu önemli ölçüde artmıştır ve FYO, PUT ve VERA veri kümeleri üzerinde sırasıyla %99.06, %99.8 ve %99.26 başarı elde edilmiştir.

Keywords

Biyometri, Avuç içi damarı, VGG16, VGG19, AlexNet, Karar-Seviyesi Kaynaşımı.

Supporting Institution

Yoktur

Project Number

Yoktur

Thanks

Yoktur

A Palm Vein Recognition Approach by Multiple Convolutional Neural Network Models

Abstract

A palm vein recognition system is proposed in this paper. The efficiency of three convolutional neural network models (VGG16, VGG19 and AlexNet) in palm vein biometrics is compared and then this study proposes to fuse them with Decision-Level Fusion. These models employ the use of high number of filters during training which leads to very high computation time, therefore, the filters are reduced in this study to drastically reduce computation time while maintaining the efficiency of the models. The proposed method is tested on three datasets secured from FYO, PUT and VERA databases. The proposed system significantly increases the accuracy of the system in comparison with the individual models and achieves 99.06 %, 99.83 % and 99.26 % on FYO, PUT and VERA datasets, respectively.

Keywords

Biometrics, Palm vein, VGG16, VGG19, AlexNet, Decision-Level Fusion.

Project Number

Yoktur

References

• Toygar, O., Babalola, F. & Bitirim, Y. (2020). FYO: A Novel Multimodal Vein Database With Palmar, Dorsal and Wrist Biometrics. IEEE Access, 8, pp.82461-82470. https://doi.org/10.1109/ACCESS.2020.2991475.
• Simonyan, K. & Zisserman A. (2015). Very Deep Convolutional Networks for Large-scale Image Recognition. In Int. Conf. on Learning Representations (ICLR), San Diego, CA, USA, pp. 1–14.
• Ha, I., Kim, H., Park, S. & Kim, H. (2018). Image retrieval using BIM and features from pretrained VGG network for indoor localization. Building and Environment, 140, pp.23-31. https://doi.org/10.1016/j.buildenv.2018.05.026.
• Krizhevsky, A., Sutskever, I. & Hinton, G. (2017). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), pp.84-90. https://doi.org/ 10.1145/3065386.
• Kabaciński, R. & Kowalski, M. (2011). Vein pattern database and benchmark results. Electronics Letters, 47(20), p.1127.
• Tome, P. & Marcel, S. (2015). On the vulnerability of palm vein recognition to spoofing attacks. In: Proceedings of 2015 International Conference on Biometrics, ICB, pp. 319-325, https://doi.org/10.1109/ICB.2015.7139056.
• Sharma, S., Dubey, S., Singh, S., Saxena, R. & Singh, R. (2015). Identity verification using shape and geometry of human hands. Expert Systems with Applications, 42(2), pp.821-832. https://doi.org/10.1016/j.eswa.2014.08.052.
• Sidiropoulos, G., Kiratsa, P., Chatzipetrou, P. & Papakostas, G. (2021). Feature Extraction for Finger-Vein-Based Identity Recognition. Journal of Imaging, 7(5), p.89. https://doi.org/10.3390/jimaging7050089.
• Nadiya, K. & Gopi, V. P. (2020). Dorsal Hand Vein Biometric Recognition Based on Orientation of Local Binary Pattern. 2020 IEEE-HYDCON, pp. 1-6, https://doi.org/10.1109/HYDCON48903.2020.9242879.
• Babalola, F., Bitirim, Y. & Toygar, Ö. (2020). Palm vein recognition through fusion of texture-based and CNN-based methods. Signal, Image and Video Processing, 15(3), pp.459-466.
• Zhang, L., Cheng, Z., Shen, Y. & Wang, D. (2018). Palmprint and Palmvein Recognition Based on DCNN and A New Large-Scale Contactless Palmvein Dataset. Symmetry, 10(4), p.78.
• Wu, K., Lee, J., Lo, T., Chang, K. & Chang, C. (2013). A secure palm vein recognition system. Journal of Systems and Software, 86(11), pp.2870-2876. https://doi.org/ 10.1016/j.jss.2013.06.065.
• Wang, P. & Sun, D. (2016). A research on palm vein recognition. 2016 IEEE 13th International Conference on Signal Processing (ICSP), pp. 1347-1351, https://doi.org/10.1109/ICSP.2016.7878046.
• Prasanthi, B.V., Hussain, S. M., Kanakam, P. & Chakravarthy, A. (2015). Palm Vein Biometric Technology: An Approach to Upgrade Security in ATM Transactions. International Journal of Computer Applications. 112, pp. 975-8887, https://doi.org/10.5120/19691-1440.
• Watanabe, M., Endoh, T., Shiohara, M., & Sasaki, S. (2005). Palm vein authentication technology and its applications. In: Proceedings of the Biometric Consortium Conference, pp. 37-38.
• Lee, J. (2012). A novel biometric system based on palm vein image. Pattern Recognition Letters, 33(12), pp.1520-1528.
• Han, W. & Lee, J. (2012). Palm vein recognition using adaptive Gabor filter. Expert Systems with Applications, 39(18), pp.13225-13234.
• Shah, G., Shirke, S., Sawant, S. & Dandawate, Y. (2015). Palm vein pattern-based biometric recognition system. International Journal of Computer Applications in Technology, 51(2), p.105.
• Athale, S., Patil, D., Deshpande, P. & Dandawate, Y. (2015). Hardware Implementation of Palm Vein Biometric Modality for Access Control in Multilayered Security System. Procedia Computer Science, 58, pp.492-498.