DETEKSI JUMLAH KENDARAAN YANG MASUK DAN KELUAR JALAN TOL MENGGUNAKAN YOLOv5 DAN DEEP SORT
DOI:
https://doi.org/10.31539/fetaqm71Abstract
Perkembangan teknologi computer vision memungkinkan proses pemantauan lalu lintas dilakukan secara ototmatis melalui deteksi, pelacakan, dan perhitungan kendaraan. Penelitian ini bertujuan untuk membangun sistem deteksi dan pelacakan kendaraan menggunakan algortima YOLOv5 dan Deep SORT pada video lalu lintas jalan tol. Dataset yang digunakan diperoleh dari proses pengambilan gambar dan video secara mandiri, kemudian dilakukan anotasi menggunakan Roboflow dengan tiga kelas kendaraan yaitu car, bus, truck. Dataset terdiri dari 326 objek kendaraan yaitu meliputi 242 mobil, 70 truk, dan 14 bis. Model YOLOv5s dilatih menggunakan 100 epoch dengan ukuran citra 640x640 piksel dan batch size 16. Hasil pengujian menunjukan bahwa model memperoleh nilai precision sebesar 94,7%, recall sebesar 92,6% mAP@0.5 sebesar 98,1% dan mAP@0.5:0.95 sebesar 66,9%. Selanjutnya, hasil deteksi diintegrasikan dengan algoritma Deep SORT untuk memberikan identitas unik pada setiap kendaraan sehingga pergerakannya dapat dilacak secara konsisten antar frame. Pengujian dilakukan menggunakan video lalu lintas berdurasi 3 menit 38 detik. Hasil penghitungan menunjukan terdapat 16 mobil masuk dan 32 mobil keluar, 2 bus masuk tanpa kendaraan bus keluar, serta 7 truk keluar tanpa ditemukan truk yang masuk. Berdasarkan hasil tersebut, kombinasi YOLOv5 dan Deep SORT mampu melakukan deteksi, pelacakan, serta penghitungan kendaraan secara otomatis dengan hasil yang baik pada kondisi lalu lintas yang diamati.
Kata Kunci : YOLOv5, Deep SORT, deteksi kendaraan, pelacakan objek, counting line, jalan tol.
References
[1] J. Tian, N. Ma, Y. Zhou, Y. Lv, W. Chi, and L. Sun, “An Efficient End-to-End Lightweight Object Detection Method Based on YOLOv5 for Intelligent Sweeping Robots An Efficient End-to-End Lightweight Obj ect Detection Method B ased on YOLOv5 for Intelligent Sweeping Robots,” IECON 2023- 49th Annu. Conf. IEEE Ind. Electron. Soc., no. March 2024, pp. 1–8, 2023, doi: 10.1109/IECON51785.2023.10311823.
[2] S. Kumar, S. K. Singh, S. Varshney, S. Singh, and P. Kumar, “Fusion of Deep Sort and Yolov5 for Effective Vehicle Detection and Tracking Scheme in Real-Time Traffic Management Sustainable System,” 2023.
[3] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” 2020, [Online]. Available: http://arxiv.org/abs/2004.10934
[4] C. Series, “Multiple Object Tracking Based on YOLOv5 and Optimized DeepSORT Algorithm Multiple Object Tracking Based on YOLOv5 and Optimized DeepSORT Algorithm”, doi: 10.1088/1742-6596/2547/1/012022.
[5] A. J. M. Plompen et al., “The joint evaluated fission and fusion nuclear data library , JEFF-3 . 3,” Eur. Phys. J. A, vol. 56, no. 7, pp. 1–108, 2020, doi: 10.1140/epja/s10050-020-00141-9.
[6] Y. Zhang, Z. Guo, J. Wu, Y. Tian, H. Tang, and X. Guo, “Real-Time Vehicle Detection Based on Improved YOLO v5,” 2022.
[7] H. Wibowo and I. S. Sitanggang, “Large-Scale Oil Palm Trees Detection from High-Resolution Remote Sensing Images Using Deep Learning,” 2022.
[8] Z. B. Khabibullah, “Tracking Socer Player Based on Deepsort Algorithm with YOLOV8 FrameWork,” vol. 9, no. 1, 2025.
[9] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, “Edinburgh Research Explorer The PASCAL Visual Object Classes ( VOC ) Challenge The P ASCAL Visual Object Classes ( VOC ) Challenge,” vol. 88, no. 2, 2010, doi: 10.1007/s11263-009-0275-4.
[10] N. J. Hayati, D. Singasatia, and M. R. Muttaqin, “Object Tracking Menggunakan Algoritma You Only Look Once (YOLO)v8 untuk Menghitung Kendaraan,” Komputa J. Ilm. Komput. dan Inform., vol. 12, no. 2, pp. 91–99, 2023, doi: 10.34010/komputa.v12i2.10654.
[11] N. Wojke, A. Bewley, and D. Paulus, “Simple online and realtime tracking with a deep association metric,” Proc. - Int. Conf. Image Process. ICIP, vol. 2017-Septe, pp. 3645–3649, 2017, doi: 10.1109/ICIP.2017.8296962.
[12] B. A. Jamiu, A. T. Oladele, M. Issa, M. Asaduzzaman, and A. Bari, “Advanced Vehicle Tracking System : Leveraging Gabor Feature Extraction and YOLOv5-Deep SORT for Real-Time Multi- Class Traffic Monitoring .,” no. January, 2025, doi: 10.35629/5252-0701217231.
[13] R. Khanam and M. Hussain, “W yolo 5: a,” pp. 3–10, 2024.
[14] Z. Huang, L. Li, G. C. Krizek, and L. Sun, “Research on Traffic Sign Detection Based on Improved YOLOv8,” J. Comput. Commun., vol. 11, no. 07, pp. 226–232, 2023, doi: 10.4236/jcc.2023.117014.
[15] M. Rahman et al., “Real-Time Vehicle Type Detection and Counting for Emission Pollution Monitoring and Traffic Violation Identification,” vol. 9, no. 2, pp. 959–976, 2025.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Stevanus Januar Latuluma, Magdalena A. Ineke Pakereng
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
