Sistem Deteksi Pejalan Kaki Pada Kendaraan Menggunakan Sistem Pengolahan Citra
DOI:
https://doi.org/10.29407/xx1jar27Keywords:
Self Driving, YOLOv8, Mobil, Kendaraan Otonom, Deteksi Objek, Data Sintetis,, Domain GapAbstract
Pengembangan sistem kendaraan otonom sangat bergantung pada kemampuan deteksi objek berbasis computer vision untuk mengenali lingkungan sekitar secara akurat dan real-time. Salah satu tantangan utama dalam pengembangan sistem ini adalah keterbatasan data dunia nyata (real-world data) yang sulit diperoleh serta perbedaan karakteristik visual antara data sintetis dan data nyata yang menyebabkan terjadinya domain gap.Penelitian ini bertujuan untuk menganalisis efektivitas pendekatan synthetic-to-real domain adaptation dalam meningkatkan performa model deteksi objek, khususnya objek pejalan kaki, pada sistem kendaraan otonom. Metode yang digunakan adalah algoritma YOLO (You Only Look Once) versi YOLOv8, yang bekerja dengan melakukan prediksi lokasi bounding box, tingkat kepercayaan, dan kelas objek secara simultan dalam satu tahap (single-stage detector). Proses deteksi dimulai dari citra masukan yang melalui tahapan preprocessing berupa resizing, normalisasi, dan augmentasi data, sebelum dilakukan inferensi oleh model YOLOv8 untuk mengenali objek pejalan kaki secara real-time.Tahapan penelitian meliputi pengumpulan dataset sintetis dan data real-time, preprocessing citra, pelatihan model, serta evaluasi performa menggunakan metrik mean Average Precision (mAP), precision, recall, dan kecepatan inferensi (Frames Per Second). Hasil penelitian menunjukkan bahwa penggunaan data campuran mampu meningkatkan kemampuan generalisasi model pada kondisi dunia nyata, dengan nilai precision sebesar 0,85, recall sebesar 0,82, serta mAP@0,5 sebesar 0,88. Hasil ini menunjukkan bahwa pendekatan synthetic-to-real domain adaptation menghasilkan performa deteksi yang lebih stabil dibandingkan pelatihan menggunakan satu domain data saja. Penelitian ini diharapkan dapat menjadi dasar pengembangan sistem deteksi objek yang lebih adaptif dan aplikatif pada kendaraan otonom berbasis kamera atau perangkat Internet of Things (IoT).
Downloads
References
[1] C. Bhardwaj, “Artificial Intelligence in Autonomous Vehicle Systems,” International Journal of Artificial Intelligence Research, vol. 9, no. 2, pp. 45–56, 2025.
[2] D. Dong, Y. Li, and Z. Wang, “Computer Vision Techniques for Autonomous Driving: A Review,” IEEE Access, vol. 13, pp. 112340–112355, 2025.
[3] A. Bansal, “Vision-Based Object Detection for Autonomous Vehicles,” Journal of Intelligent Transportation Systems, vol. 27, no. 1, pp. 15–28, 2023.
[4] Y. Cholandha, “Perkembangan Teknologi Kendaraan Otonom di Industri Otomotif Global,” Jurnal Teknologi Transportasi, vol. 6, no. 2, pp. 87–95, 2024.
[5] T. Lynn, “Synthetic Data Generation for Computer Vision Applications,” Machine Learning Advances, vol. 4, no. 1, pp. 22–35, 2023.
[6] J. Carlson, A. Shrivastava, and P. Sermanet, “Understanding the Domain Gap in Synthetic-to-Real Visual Learning,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1234–1243, 2018.
[7] Politecnico di Torino, Image Processing for Self-Driving Cars, Master Thesis, Torino, Italy, 2022.
[8] S. Donthi, R. Garg, and A. Patel, “Self-Driving Car to Drive Autonomously Using Image Processing and Deep Learning,” International Research Journal of Engineering and Technology (IRJET), vol. 9, no. 6, pp. 1180–1186, 2022.
[9] M. Dastres and H. Soori, “Image Processing Techniques and Applications: A Review,” Journal of Advanced Research in Computer Science, vol. 12, no. 3, pp. 45–58, 2021.
[10] M. Zadobrischi and M. Dimian, “Object Detection Performance under Domain Shift Conditions,” Remote Sensing, vol. 13, no. 9, pp. 1–18, 2021.
[11] J. Redmon and A. Farhadi, “YOLO: You Only Look Once—Unified, Real-Time Object Detection,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 779–788, 2018.
[12] Ultralytics, “YOLOv8 Documentation,” 2023. [Online]. Available: Ultralytics Official Documentation.
[13] ASTB, “Real-Time Object Detection Systems for Autonomous Vehicles,” Automotive Sensor Technology Bulletin, vol. 5, no. 1, pp. 33–40, 2022.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Zhico Putra Hery Prianto
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Copyright on any article is retained by the author(s).
- The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal’s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
- The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License
