Deteksi dan Penghitungan Speed Bump Berbasis YOLOv8 dan DeepSORT untuk Sistem Keselamatan Berkendara

Immawan Wicaksono; Hilman Arif Rahmawan; Ali Rizal Chaidir; Andrita Ceriana Eska; Moh Asnoer Laagu

doi:10.30872/jurti.v10i3.26828

Deteksi dan Penghitungan Speed Bump Berbasis YOLOv8 dan DeepSORT untuk Sistem Keselamatan Berkendara

Immawan Wicaksono, Hilman Arif Rahmawan, Ali Rizal Chaidir, Andrita Ceriana Eska, Moh Asnoer Laagu

Abstract

Keselamatan berkendara merupakan aspek penting dalam sistem transportasi, terutama pada jalan yang dilengkapi speed bump sebagai pembatas kecepatan. Speed bump yang tidak terlihat jelas dapat menimbulkan ketidaknyamanan, kerusakan kendaraan, dan potensi kecelakaan. Penelitian ini mengembangkan sistem deteksi dan penghitungan speed bump berbasis You Only Look Once version 8 (YOLOv8) yang dikombinasikan dengan DeepSORT dan garis virtual pada video rekaman kamera smartphone. Dataset yang digunakan terdiri atas 1.000 citra speed bump yang dianotasi secara manual menggunakan bounding box dan dibagi menjadi data latih, validasi, dan uji dengan rasio 70:20:10. Model dilatih pada Google Colab selama 100 epoch dengan ukuran citra 640 x 640 piksel dan batch size 32. Evaluasi dilakukan menggunakan precision, recall, mAP, akurasi deteksi, akurasi counting, dan frame per second (FPS). Hasil validasi menunjukkan precision 99,85%, recall 100%, mAP@0.5 sebesar 99,50%, dan mAP@0.5:0.95 sebesar 64,16%. Pada pengujian video lapangan, confidence threshold 0,5 memberikan rata-rata akurasi deteksi tertinggi sebesar 85,50%, sedangkan threshold 0,6 dan 0,7 masing-masing menghasilkan 63,64% dan 37,01%. Kecepatan pemrosesan sistem berada pada rentang 25-42 FPS. Hasil penelitian menunjukkan bahwa YOLOv8 mampu mendeteksi speed bump dengan baik pada kondisi pencahayaan memadai, namun akurasi penghitungan masih dipengaruhi oleh double counting akibat ketidakstabilan ID tracking.

Keywords

DeepSORT; deteksi objek; keselamatan jalan; speed bump; YOLOv8

Full Text:

PDF

References

Kementerian Perhubungan Republik Indonesia, 2021, Peraturan Menteri Perhubungan Nomor 14 Tahun 2021 tentang Alat Pengendali dan Pengaman Pengguna Jalan, Kementerian Perhubungan Republik Indonesia, Jakarta.

Peralta-López, J.-E., Morales-Viscaya, J.-A., Lázaro-Mata, D., Villaseñor-Aguilar, M.-J., Prado-Olivarez, J., Pérez-Pinal, F.-J., Padilla-Medina, J.-A., Martínez-Nolasco, J.-J., and Barranco-Gutiérrez, A.-I., 2023, Speed bump and Pothole Detection Using Deep Neural Network with Images Captured through ZED Camera, Applied Sciences, Vol. 13, No. 14, Art. no. 8349, doi: https://doi.org/10.3390/app13148349.

Wang, R., Luo, X., Ye, Q., Jiang, Y., and Liu, W., 2024, Research on Visual Perception of Speed bumps for Intelligent Connected Vehicles Based on Lightweight FPNet, Sensors, Vol. 24, No. 7, Art. no. 2130, doi: https://doi.org/10.3390/s24072130.

Xiang, X., Chen, X., Tang, X., Luo, Q., and Ding, Y., 2024, Research on Detection of Multiple Types of Speed bump Defects Based on CRSCCG-YOLOv5s, IEEE Access, Vol. 12, pp. 116786–116800.

Nugraha, E. I., and Pranoto, Y. M., 2021, Deteksi Polisi Tidur pada Jalan Menggunakan Analisis Blob dan Konvolusional Neural Network, Journal of Intelligent Systems and Computation, pp. 16–22.

Taufiqurrahman, T., Hadi, A. P., and Siregar, R. E., 2024, Evaluasi Performa YOLOv8 dalam Deteksi Objek di Depan Kendaraan dengan Variasi Kondisi Lingkungan, Jurnal Minfo Polgan, Vol. 13, No. 2, pp. 1755–1773.

Santoso, A. D., Nugroho, J. B., and Setyaningtyas, E. P., 2024, Deteksi Jenis Kendaraan di Jalan Menggunakan OpenCV dan YOLOv4, Jurnal AI dan SPK, Vol. 2, No. 2, pp. 127–131.

Megantara, N. A., and Utami, E., 2025, Object Detection Using YOLOv8: A Systematic Review, Sistemasi: Jurnal Sistem Informasi, Vol. 14, No. 3.

Jocher, G., Chaurasia, A., and Qiu, J., 2023, Ultralytics YOLO, Software documentation, URL: https://docs.ultralytics.com/.

Zhao, R., Tang, S. H., Supeni, E. E. B., Rahim, S. A., and Fan, L., 2024, Z-YOLOv8s-Based Approach for Road Object Recognition in Complex Traffic Scenarios, Alexandria Engineering Journal, Vol. 106, pp. 298–311, doi: https://doi.org/10.1016/j.aej.2024.07.011.

Khalili, B., and Smyth, A. W., 2024, SOD-YOLOv8: Enhancing YOLOv8 for Small Object Detection in Traffic Scenes, arXiv preprint arXiv:2408.04786, URL: https://arxiv.org/abs/2408.04786.

Zhu, Y., Wang, Y., An, Y., Yang, H., and Pan, Y., 2024, Real-Time Vehicle Detection and Urban Traffic Behavior Analysis Based on UAV Traffic Videos on Mobile Devices, arXiv preprint arXiv:2402.16246, URL: https://arxiv.org/abs/2402.16246.

Schneidereit, T., Gohrenz, S., and Breuß, M., 2025, Object Detection Characteristics in a Learning Factory Environment Using YOLOv8, arXiv preprint arXiv:2503.10356, URL: https://arxiv.org/abs/2503.10356.

Cahyani, P. A., Mardiana, M., Wintoro, P. B., and Muhammad, M. A., 2024, Vehicle Counting System Using YOLOv5 and DeepSORT Algorithms, Jurnal Teknik Informatika dan Sistem Informasi, Vol. 10, No. 1, pp. 86–99, doi: https://doi.org/10.28932/jutisi.v10i1.7519.

Kumar, S., Singh, S. K., Varshney, S., Singh, S., Kumar, P., Kim, B.-G., and Ra, I.-H., 2023, Fusion of Deep SORT and YOLOv5 for Effective Vehicle Detection and Tracking Scheme in Real-Time Traffic Management Sustainable System, Sustainability, Vol. 15, No. 24, Art. no. 16869, doi: https://doi.org/10.3390/su152416869.

Yang, F., Zhang, X., and Liu, B., 2022, Video Object Tracking Based on YOLOv7 and DeepSORT, arXiv preprint arXiv:2207.12202, URL: https://arxiv.org/abs/2207.12202.

Durve, M., Orsini, S., Tiribocchi, A., Montessori, A., Tucny, J.-M., Lauricella, M., Camposeo, A., Pisignano, D., and Succi, S., 2023, Benchmarking YOLOv5 and YOLOv7 Models with DeepSORT for Droplet Tracking Applications, arXiv preprint arXiv:2301.08189, URL: https://arxiv.org/abs/2301.08189.

Wang, C.-Y., Bochkovskiy, A., and Liao, H.-Y. M., 2022, YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors, arXiv preprint arXiv:2207.02696, URL: https://arxiv.org/abs/2207.02696.

Liang, C., Fu, J., Luo, L., and Yu, M., 2025, SBP-YOLO: A Lightweight Real-Time Model for Detecting Speed bumps and Potholes, arXiv preprint arXiv:2508.01339, URL: https://arxiv.org/abs/2508.01339.

Antariksa, G., Chakraborty, R., Somvanshi, S., Das, S., Jalayer, M., Patel, D. R., and Mills, D., 2025, Comparative Analysis of Advanced AI-Based Object Detection Models for Pavement Marking Quality Assessment during Daytime, arXiv preprint arXiv:2503.11008, URL: https://arxiv.org/abs/2503.11008.

DOI: http://dx.doi.org/10.30872/jurti.v10i3.26828

Refbacks

There are currently no refbacks.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Alamat Redaksi :
Program Studi Informatika
Fakultas Teknik
Jl. Sambaliung No. 9 Kampus Gunung Kelua Samarinda 75119 - Kalimantan Timur
e-mail : jurti.unmul@fkti.unmul.ac.id
Url : http://e-journals.unmul.ac.id/index.php/INF
Contact Person : Medi Taruk [081528400126]

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.