Pengembangan Model Computer Vision dengan Arsitektur YOLOv12 untuk Deteksi Utilitas Bawah Tanah pada Data Ground Penetrating Radar
DOI:
https://doi.org/10.29303/goescienceed.v6i4.1380Keywords:
GPR, Utilitas, Deep Learning, YOLOv12Abstract
Abstract: Perkembangan infrastruktur perkotaan yang pesat menuntut metode pemetaan utilitas bawah tanah yang akurat dan efisien untuk mitigasi risiko selama konstruksi. Keterbatasan interpretasi manual data Ground Penetrating Radar (GPR), yang bersifat subjektif, lambat, dan memerlukan keahlian tinggi, menjadi hambatan utama dalam efisiensi survei geofisika. Penelitian ini berfokus pada deteksi objek secara end-to-end berbasis deep learning yang memanfaatkan arsitektur YOLOv12 untuk otomatisasi identifikasi anomali hiperbolik yang merepresentasikan utilitas pada citra radargram. Model prediksi dilatih dan divalidasi pada dataset utilitas GPR kustom yang telah melalui tahap pra-pemrosesan dan anotasi yang cermat. Hasil pengembangan awal model menunjukkan bahwa model YOLOv12 mencapai kinerja yang cukup baik, dengan mean Average Precision (mAP0.5) sebesar 70%, precision 80% serta recall 70%. Studi ini menunjukkan bahwa arsitektur YOLOv12 bisa menjadi solusi yang kuat, akurat, dan efisien untuk otomatisasi pemetaan utilitas bawah tanah, yang berpotensi merevolusi praktik survei geofisika di bidang teknik sipil dan manajemen infrastruktur.
Keywords: GPR, Utilitas, Deep Learning, YOLOv12
References
Amaral, L. C. M., Roshan, A., & Bayat, A. (2023). Automatic detection and classification of underground objects in ground penetrating radar images using machine learning. Journal of Pipeline Systems Engineering and Practice, 14(4), 04023040. https://doi.org/10.1061/JPSEA2.PSENG-1444
Hou, F., Lei, W., Li, S., & Xi, J. (2021). Deep learning-based subsurface target detection from GPR scans. IEEE sensors journal, 21(6), 8161-8171. https://doi.org/10.1109/JSEN.2021.3050262
Hou, F., Rui, X., Fan, X., & Zhang, H. (2022). Review of GPR activities in civil infrastructures: Data analysis and applications. Remote Sensing, 14(23), 5972. https://doi.org/10.3390/rs14235972
Liu, Z., Wu, W., Gu, X., Li, S., Wang, L., & Zhang, T. (2021). Application of Combining YOLO Models and 3D GPR Images in Road Detection and Maintenance. Remote Sensing, 13(6), 1081. https://doi.org/10.3390/rs13061081
Pashoutani, S., Roudsari, M., & Zhu, J. (2025). Dual-Channel CNN-Based Framework for Automated Rebar Detection in GPR Data of Concrete Bridge Decks. Construction Materials, 5(2), 36. https://doi.org/10.3390/constrmater5020036
Wang, A., Chen, H., Liu, L., Chen, K., Lin, Z., & Han, J. (2024). Yolov10: Real-time end-to-end object detection. Advances in Neural Information Processing Systems, 37, 107984-108011.
Wang, C. Y., Yeh, I. H., & Mark Liao, H. Y. (2024, September). Yolov9: Learning what you want to learn using programmable gradient information. In European conference on computer vision (pp. 1-21). Cham: Springer Nature Switzerland.
Wang, H., Luo, J., He, Y., Cao, M., Gao, D., & Li, J. (2024). Identification of common underground pipelines by ground penetrating radar based on improved YOLOv5. Chinese Journal of Geophysics, 67(9), 3588-3604. https://doi.org/10.6038/cjg2023R0431
Yan, Y., Jiao, G., Cui, M., & Ni, L. (2025). The Lightweight Method of Ground Penetrating Radar (GPR) Hidden Defect Detection Based on SESM-YOLO. Buildings, 15(13), 2345. DOI: https://doi.org/10.3390/buildings15132345
Zhong, Y., Li, K., Mo, S., & Li, B. (2023). Research on GPR Echo Image Recognition Based on GSR-YOLOv7. Semiconductor Optoelectronics, 44(5), 767-772. https://doi.org/10.16818/j.issn1001-5868.2023061401
Zhu, X., Lyu, S., Wang, X., & Zhao, Q. (2020). Research on GPR image recognition based on deep learning. MATEC Web of Conferences, 309, 03027. https://doi.org/10.1051/matecconf/202030903027
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Jurnal Pendidikan, Sains, Geologi dan Geofisika (GeoScienceed)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
