VEHICLE COUNTING BASED ON OBJECT DETECTION USING THE YOLOV4-TINY ALGORITHM: PENGHITUNG JUMLAH KENDARAAN BERBASIS DETEKSI OBJEK MENGGUNAKAN ALGORITMA YOLOV4-TINY

Rian Setiyana; Hadi Nasbey; Haris Suhendar

doi:10.21009/03.1301.FA14

Authors

Rian Setiyana Program Studi Fisika, FMIPA, Universitas Negeri Jakarta
Hadi Nasbey Program Studi Fisika, FMIPA, Universitas Negeri Jakarta
Haris Suhendar Program Studi Fisika, FMIPA, Universitas Negeri Jakarta

DOI:

https://doi.org/10.21009/03.1301.FA14

Abstract

Traffic congestion occurs when vehicle flow is disrupted or completely halted. This situation is often encountered in urban areas, especially when there is an imbalance between the number of vehicles and the available road capacity. Compared to the rate of urban road infrastructure development, the increase in the number of vehicles has grown exponentially. From 2013 to 2017, the average annual increasein the number of vehicles was approximately 8.600.000. This has led to high levels of traffic congestion in urban areas. Therefore, this study aims to monitor traffic density by developing a vehicle counting system based on object detection. This system is designed to count the number of vehicles in video recordings. In this study, the vehicle counting system utilizes the object detection capabilities of the You Only Look Once (YOLO) algorithm, specifically YOLOv4-tiny. YOLOv4-tiny was chosen for its smaller model size and faster detection capabilities. Based on the training results, the mAP of the best YOLOv4-tiny weights reached 95.13%. Furthermore, the counting results indicate that the system successfully counted vehicles passing the auxiliary line based on their types.

References

[1] D. A. Abdurrafi, M. T. Alawiy, and B. M. Basuki, "Deteksi klasifikasi dan menghitung kendaraan berbasis algoritma You Only Look Once (YOLO) menggunakan kamera CCTV," Science Electro, pp. 1-6, 2023.

[2] S. Sahara and B. N. Nugroho, "Efektivitas penggunaan kereta listrik (KRL) Commuter Line Jabodetabek untuk mengurangi kemacetan di DKI Jakarta," Jurnal Ekonomika45, pp. 415-426, 2023.

[3] M. A. Santoso, J. Raharjo, and N. Ibrahim, "Implementasi alat pemantauan kepadatan lalu lintas," e-Proceeding of Engineering, pp. 2981-2988, 2022.

[4] F. Novaldi, I. Amrulloh, I. W. Wisesa, and M. C. Manullang, "Pendeteksian pelanggaran pada penyebrangan jalan menggunakan single-shot detector pada ESP32," Tematik: Jurnal Teknologi Informasi Komunikasi, pp. 119-127, 2022.

[5] D. Agustiani, "Implementasi machine learning dan computer vision pada pengembangan sistem otomasi klasifikasi dan perhitungan kendaraan," Seminar Nasional Dinamika Informatika 2019, pp. 16-19, 2019.

[6] A. A. Suradi, M. F. Rasyid, and Nasaruddin, "Sistem perhitungan jumlah kendaraan berbasis computer vision," Prosiding Seminar Ilmiah Sistem Informasi dan Teknologi Informasi, pp. 89- 97, 2022.

[7] T. Sutisna, A. R. Raharja, Solihin, E. Hariyadi, and V. H. Putra, "Penggunaan computer vision untuk menghitung jumlah kendaraan dengan menggunakan metode SSD (Single Shot Detector)," Innovative: Journal of Social Science Research, pp. 6060-6067, 2024.

[8] D. Indra,Herman, and F. S. Budi, "Implementasi sistem penghitung kendaraan otomatis berbasis computer vision," Komputika: Jurnal Sistem Komputer, pp. 53-62, 2023.

[9] F. Rachmawati and D. Widhyaestoeti, "Deteksi jumlah kendaraan di jalur SSA Kota Bogor menggunakan algoritma deep learning YOLO," Prosiding LPPM UIKA Bogor, pp. 360-370, 2020.

[10] R. M. Yusup, A. F. Anugrah, D. D. Muslimah, S. M. Permana, and S. Yuliani, "Pendeteksian objek menggunakan OpenCV dan metode YOLOv4-tiny untuk membantu tunanetra," Journal of Computer Science and Information Technology, pp. 59-68, 2024.