Revista ELECTRO
Vol. 44 – Año 2022
Artículo
TÍTULO
Embedded Vision System with Multiple Cameras for Face Recognition
AUTORES
Andrés Enrique Loya Domínguez, Isidro Robledo Vega, Pedro Rafael Márquez Gutiérrez, Carmen Leticia García Mata, Alberto Pacheco González
RESUMEN
Palabras Clave:
ABSTRACT
This article presents the development of an embedded vision system with multiple cameras for monitoring acce ss areas to a building and to carry on personnel identification via facial recognition. A Deep Neural Network (DNN) was trained on the VGGFace2 dataset using two different architectures, DFN-L and ResNet-50, as well as two different classification layers, fully connected and ArcFace. The behavior during training and discriminative performance of each combination of architecture and classification layer is compared to determine which is more appropriate for a continuously re-trainable network, observing that a ResNet-50 with regular fully connected classification layer fits the criteria better. This network was retrained on a dataset to identify a people which data was acquired locally, reaching 98.83% verification accuracy. This DNN was deployed on an Nvidia Jetson TX2 embedded system connected to a surveillance camera system to identify persons in the database.
Keywords: Face Recognition, Face Detection, Neural Networks, Biometric Systems, Deep Learning.
REFERENCIAS
[1] A. Jain, R. Bolle, S. Pankati, Biometrics: Personal Identification in Networked Society. Norwell, MA: Kluwer Academic Publishers, 1999.
[2] L. Masupha, T. Zuva, S. Ngwira, O. Esan. “Face recognition Techniques, their Advantages, Disadvantages and Performance Evaluation”, 2015 International Conference on Computing, Communication and Security (ICCCS)
[3] P. Viola, M. Jones, “Rapid object detection using a boosted cascade of simple features”, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Kauai, HI, USA, 8–14 December 2001
[4] K. Zhang, Z. Zhang, Z. Li and Y. Qiao, "Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks," in IEEE Signal Processing Letters, vol. 23, no. 10, pp. 1499-1503, Oct. 2016, doi: 10.1109/LSP.2016.2603342.
[5] Y. LeCun and Y. Bengio, "Convolutional networks for images, speech, and time series," in The Handbook of Brain Theory and Neural Networks, A. A. Michael, Ed. Cambridge, MA, USA: MIT Press, 1998, pp. 255–258.
[6] M.He, J. Zhang, S. Shan, M. Kan, X. Chen, “Deformable face net for pose invariant face recognition”, in Pattern Recognition, 107113. doi: 10.1016/j.patcog.2019.107 1, 2019.
[7] F. Wang, J. Cheng, W. Liu and H. Liu, "Additive Margin Softmax for Face Verification," in IEEE Signal Processing Letters, vol. 25, no. 7, pp. 926-930, July 2018, doi: 10.1109/LSP.2018.2822810.
[8] J. Deng, J. Guo, N. Xue, S. Zafeiriou, “ArcFace: Additive Angular Margin Loss for Deep Face Recognition”, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 4690-4699
[9] X. Zhang, R. Zhao, Y. Qiao, X. Wang, H. Li, “AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations”, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 10823-10832
[10] K. He, X. Zhang, S. Ren, J. Sun. "Deep Residual Learning for Image Recognition". Proceedings of the IEEE Computer Vision and Pattern Recognition, 2016. p. 770-778.
[11] Q. Cao, L. Shen, W. Xi e, O. M. Parkhi and A. Zisserman, "VGGFace2: A Dataset for Recognising Faces across Pose and Age," 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018), 2018, pp. 67-74, doi: 10.1109/FG.2018.00020.
[12] “TFRecord and tf.train.Example ”, TensorFlow Core, 2021. [Online] Available at: https://www.tensorflow.org/tutorials/load_data/tfrecord [Last Accessed: 12/22/2021]
[13] L. N. Smith, " A disciplined approach to neural network hyper-parameters: Part 1--learning rate, batch size, momentum, and weight decay”. arXiv preprint arXiv:1803.09820, 2018.
[14] “Jetson TX2 Module”, NVIDIA Developer, 2018. [Online] Available at: https://developer.nvidia.com/embedded/jetson-tx2 [Last Accessed: 11/27/2019]
CITAR COMO:
Andrés Enrique Loya Domínguez, Isidro Robledo Vega, Pedro Rafael Márquez Gutiérrez, Carmen Leticia García Mata, Alberto Pacheco González, "Embedded Vision System with Multiple Cameras for Face Recognition", Revista ELECTRO, Vol. 44, 2022, pp. 200-205.
VERSIÓN PDF