本論文設計研究一套協助人物資訊記憶的即時人臉辨識系統。本系統除了能即時正確辨識目標人物之外，輕便且能隨身攜帶也為其中的目標，故選擇則使用嵌入式系統為硬體平台，搭配支援USB介面之相機以取得目標人物的影像資料。
當使用者獲得目標影像，本論文透過OpenCV所提供基於Haar特徵分類的人臉偵測器來做人臉偵測( Face Detection )。透過此Haar特徵分類的人臉偵測器可以有93.1%的準確度獲得目標人物人臉，再透過TensorFlow Lite框架將MobileFaceNet模型部屬在嵌入式平台上進行人臉辨識 ( Face Recognition )，並取得特徵值。並將特徵值在標記後儲存於資料庫中。此特徵值可以與其他目標影像的特徵值加以比較，即可推測兩個人是否為同一人。本論文設計之系統對於人物辨識在一秒5幀圖像的辨識速度下，準確率為82%。

In this study, we design a real-time face recognition system that assists in checking if a just-met target person’s information is stored in the database of a wearable device. . The hardware platform is an embedded system, and the image of the target person is acquired through a USB camera.
When the user acquires to take the target image, the system would response if the person’s information is in the database. The implementation uses OpenCV Haar Cascades for face detection. The Haar Cascades face detector in the system has an accuracy of 93.1%. After acquiring the position and size of the target face, the system stores the face image. For face recognition, the MobileFaceNet model is attached to the embedded platform through the TensorFlow Lite framework to complete the face recognition. After putting the stored target faces into the MobileFaceNet model, the characteristic metric of the target person is obtained. By comparing the value of the metric with that of other target images in the database, the system would conclude whether there is a known person's image. The experimental result shows that the proposed system achieves 82% accuracy in face recognition at 5 fps.

[1] P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features", Proc. Computer vision and Pattern Recognition 2001., pp. 511-518, 2001.
[2] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection", 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition., pp. 886-893, 2005.
[3] T. Ojala, M. Pietikäinen and T. Mäenpää, "Multiresolution gray-scale and rotation invariant texture classification with localbinary patterns", IEEE Transactions on Pattern Analysis and Machine Intelligence., vol. 24, no. 7, pp. 971-987, 2002.
[4] H. Li, Z. Lin, X. Shen, J. Brandt and G. Hua, "A convolutional neural network cascade for face detection", Proceedings of the IEEE conference on computer vision and pattern recognition., pp. 5325-5334, 2015.
[5] J. Xiang and G. Zhu, "Joint face detection and facial expression recognition with MTCNN", Proceedings of 4th International Conference on Information Science and Control Engineering., pp. 424-427, Jul. 2017.
[6] G. Bradski and A. Kaehler, Learning OpenCV: Computer vision with the OpenCV library, O'Reilly Media, Inc., 2008.
[7] M. A. Turk and A. P. Pentland, "Face recognition using eigenfaces", Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition., pp. 586-591, 1991.
[8] T. Ahonen, A. Hadid and M. Pietikainen, "Face description with local binary patterns: Application to face recognition", Proceedings of IEEE Transactions on Pattern Analysis and Machine Intelligence., vol. 28, no. 12, pp. 2037-2041, Dec. 2006.
[9] Y. Taigman, M. Yang and M. Ranzato, "DeepFace: Closing the gap to human-level performance in face verification", Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Vis. Pattern Recognit., pp. 1701-1708, 2014.
[10] F. Schroff, D. Kalenichenko and J. Philbin, "FaceNet: A unified embedding for face recognition and clustering", Proceedings of IEEE Conference on Computer Vision and Pattern Recognition., pp. 815-823, Jun. 2015.
[11] S. Chen, Y. Liu, X. Gao and Z. Han, "MobileFaceNets: Efficient cnns for accurate real-time face verification on mobile devices", Chinese Conference on Biometric Recognition., pp. 428-438, 2018.
[12] J. Lee, Y. Su and C. Shen, "A comparative study of wireless protocols: Bluetooth UWB ZigBee and Wi-Fi", The 33rd Annual Conference of the IEEE Industrial Electronics Society., pp. 46-51, Nov 5–8, 2007.
[13] E. Mackensen, M. Lai and T. M. Wendt, "Bluetooth low energy (BLE) based wireless sensors", Proc. IEEE Sensors., pp. 1-4, Oct. 2012.
[14] 吳其軒 , Design and implement a face recognition system combining embedded device and mobile phone application , 碩士論文,國立成功大學工程科學系. 2022
[15] 邱子懿 , Implementation of age prediction and classification using facial recognition neural networks applicable to mobile devices , 碩士論文,國立成功大學工程科學系. 2022
[16] Sandeep Mistry, bleno, https://github.com/sandeepmistry/bleno, 2017. Last retrieved Sep, 2022
[17] L. Wolf, T. Hassner and I. Maoz, "Face recognition in unconstrained videos with matched background similarity", 2011 IEEE Conference on Computer Vision and Pattern Recognition., pp. 529-534, 2011.
[18] C. H. Tseng, Deepface, https://chtseng.wordpress.com/2021/12/04/deepface,2021. Last retrieved Sep, 2022