Implementasi Face Recognition Berbasis Deep Neural Network Sebagai Sistem Kendali Pada Quadcopter
DOI:
https://doi.org/10.34818/eoe.v9i6.18960Abstrak
Abstrak—Artifcial Intelligence (AI) adalah bidang yang dikembangkan untuk mempelajari dan menyerupai kecerdasan manusia ke dalam sebuah sistem komputer. Salah satu hasil dari pengembangan AI khususnya pada bidang deep neural network adalah teknologi face recognition. Face recognition merupakan suatu sistem yang digunakan untuk mengenali wajah manusia yang terdapat pada suatu citra. Tugas akhir ini dilakukan implementasi teknologi face recognition terhadap unmanned aerial vehicle (UAV) untuk beberapa kepentingan seperti pengawasan, pencarian orang, dan pemantauan jarak jauh. Dalam tugas akhir ini penerapan face recognition diimplementasikan menggunakan algoritma YOLOv5 sebagai object detection serta pembuatan algoritma object tracking. Untuk mengetahui performansi terbaik, penelitian ini akan dilakukan pengujian terhadap 3 model, yaitu YOLOv5n, YOLOv5s dan YOLOv5m. Untuk mencari model terbaik akan dievaluasi menggunakan parameter Mean Average Precision (mAP), jarak efektif deteksi dan kecepatan inferensi model. Hasil perbandingan menunjukkan model YOLOv5n memiliki kecepatan inferensi terbaik yaitu 5ms. Namun, model ini memiliki nilai mAP terendah yaitu 79,8%. Sedangkan, mAP tertinggi dengan nilai 84,8% dicapai oleh model YOLOv5m dengan kecepatan inferensi sebesar 13ms. Model YOLOv5s memiliki selisih kecepatan inferensi yang kecil dengan YOLOv5n dengan 6ms dan nilai mAP sebesar 82,4%. Kata Kunci— Face Recognition, Unmanned Aerial Vehicle, YOLOv5, Deep Neural Network.Unduhan
Referensi
[1] H.-J. Hsu and K.-T. Chen, “Face recognition on drones,” 05 2015, pp. 39–44.
[2] I. Goodfellow, Y. Bengio, and A. Courville, Deep learning. MIT press, 2016.
[3] R. Ranjan, A. Bansal, J. Zheng, H. Xu, J. Gleason, B. Lu, A. Nanduri, J.-C. Chen, C. D. Castillo, and R. Chellappa, “A fast and accurate system for face detection, identifcation, and verifcation,” IEEE Transactions on Biometrics, Behavior, and Identity Science, vol. 1, no. 2, pp. 82–96, 2019.
[4] H.-J. Hsu and K.-T. Chen, “Face recognition on drones,” 05 2015, pp. 39–44.
[5] Y. Fan, Y. Luo, and X. Chen, “Research on face recognition technology based on improved YOLO deep convolution neural network,” Journal of Physics: Conference Series, vol. 1982, no. 1, p. 012010, jul 2021. [Online]. Available: https://doi.org/10.1088/1742-6596/1982/1/012010
[6] Droneomega. What is a quadcopter explained thoroughly. [Online]. Available: https://droneomega.com/what-is-a-quadcopter/
[7] Raydiant. Everything about face tracking. [Online]. Available: https: //sightcorp.com/knowledge-base/face-tracking/#:∼:text=Face%20Tracking% 20Technology%20detects%20and,be%20used%20online%20or%20offline.
[8] P. Andono and T. Sutojo, Pengolahan Citra Digital. Penerbit Andi. [Online]. Available: ”https://books.google.co.id/books?id=zUJRDwAAQBAJ”
[9] T. Contributor. convolutional neural network. [Online]. Available: https: //searchenterpriseai.techtarget.com/defnition/convolutional-neural-network
[10] T. H. Afridi. A multimodal memes classifcation: A survey and open research issues. [Online]. Available: https://www.researchgate.net/figure/ A-generic-CNN-Architecture fg1 344294512
[11] J. Du, “Understanding of object detection based on cnn family and yolo,” Journal of Physics: Conference Series, vol. 1004, p. 012029, 04 2018.
[12] F. E. Ramadhan, “Penerapan image classifcation dengan pre-trained model mobilenet dalam client-side machine learning,” B.S. thesis, Fakultas Sains dan Teknologi Universitas Islam Negeri Syarif Hidayatullah . . . , 2020
[13] M. Maithani. Guide to yolov5 for real-time object detection. [Online]. Available: https://analyticsindiamag.com/yolov5/
[14] X. Zhu, S. Lyu, X. Wang, and Q. Zhao, “Tph-yolov5: Improved yolov5 based on transformer prediction head for object detection on drone-captured scenarios,” CoRR, vol. abs/2108.11539, 2021. [Online]. Available: https://arxiv.org/abs/2108.11539
[15] G. Jocher. Yolov5. [Online]. Available: https://github.com/ultralytics/yolov5
[16] B. Yan, P. Fan, X. Lei, Z. Liu, and F. Yang, “A real-time apple targets detection method for picking robot based on improved yolov5,” Remote Sensing, vol. 13, no. 9, 2021. [Online]. Available: https: //www.mdpi.com/2072-4292/13/9/1619
[17] python. What is python? executive summary. [Online]. Available: https://www.python.org/doc/essays/blurb/
[18] M. G. Meysam Vakili and M. Rezae, “Performance analysis and comparison of machine and deep learning algorithms for iot data classifcation,” Materials Today: Proceedings, 01 2020.
[19] V. Meel. How to analyze the performance of machine learning models. [Online]. https://viso.ai/deep-learning/analyzing-machine-learning-model-performance-strategies/
[20] A. chaurasia. How are your yolov5 models doing? [Online]. Available: https://wandb.ai/ayush/yoloV5/reports/ How-are-your-YOLOv5-models-doing---VmlldzoyNjM3MTY



