Sistem Deteksi Tingkat Kerusakan Dinding Berbasis YOLOv8n-OBB Untuk Mendukung Pemeliharaan
DOI:
https://doi.org/10.29407/kv2xt982Keywords:
Berjamur, Dinding, Prediksi, Retak, YOLOv8n - OBBAbstract
Kekuatan struktural bangunan sangat vital bagi keselamatan, namun inspeksi visual manual sering kali bersifat subjektif dan tidak efisien dalam mendeteksi kerusakan dini seperti retak dan jamur. Penelitian ini bertujuan mengatasi masalah tersebut dengan mengusulkan sistem pemantauan kesehatan struktural otomatis berbasis deep learning. Metode yang digunakan adalah pendekatan kuantitatif eksperimental dengan mengimplementasikan arsitektur YOLOv8n-OBB (Oriented Bounding Box). Varian 'nano' dipilih untuk mencapai keseimbangan optimal antara kecepatan inferensi real-time dan efisiensi waktu pemrosesan gambar. Model dilatih menggunakan 15.553 citra dinding yang dianotasi secara manual dalam format koordinat delapan titik. Hasil evaluasi menunjukkan performa impresif dengan nilai mean Average Precision (mAP@50) mencapai 90,7%, precision 87,2%, dan recall 84,9%. Secara teknis, model mampu memproses citra dengan kecepatan 15,08 ms, memungkinkan deteksi instan pada aplikasi berbasis web Streamlit yang juga menyediakan fitur estimasi luas area dan skor keparahan. Keberhasilan ini memvalidasi bahwa penggunaan YOLOv8n-OBB secara signifikan meningkatkan presisi lokalisasi kerusakan dibandingkan bounding box standar. Implementasi sistem ini memberikan kontribusi penting dalam mengotomatisasi inspeksi bangunan, mengurangi biaya pemeliharaan, serta meminimalisir risiko kegagalan struktural melalui deteksi dini yang objektif dan aplikatif bagi masyarakat luas.
Downloads
References
[1] G. Wang dan J. Ke, “Literature Review on the Structural Health Monitoring (SHM) of Sustainable Civil Infrastructure: An Analysis of Influencing Factors in the Implementation,” Build. 2024 Vol 14 Page 402, vol. 14, no. 2, hlm. 402, Feb 2024, doi: 10.3390/BUILDINGS14020402.
[2] C. Karakostas dkk., “Seismic assessment of bridges through structural health monitoring: a state-of-the-art review,” Bull. Earthq. Eng. 2023 223, vol. 22, no. 3, hlm. 1309–1357, Nov 2023, doi: 10.1007/S10518-023-01819-3.
[3] H. Abou-Ibrahim dan F. Hamzeh, “Understanding Stakeholders’ impact on design workflow dynamics using agent-based modeling,” Autom. Constr., vol. 138, hlm. 104254, Jun 2022, doi: 10.1016/J.AUTCON.2022.104254.
[4] J. Jia dan Y. Li, “Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends,” Sensors, vol. 23, no. 21, hlm. 8824, Okt 2023, doi: 10.3390/S23218824/S1.
[5] F. Poli, M. F. Bado, A. Verzobio, dan D. Zonta, “Bridge structural safety assessment: a novel solution to uncertainty in the inspection practice,” Struct. Infrastruct. Eng., vol. 21, no. 3, hlm. 421–435, 2023, doi: 10.1080/15732479.2023.2211956;WGROUP:STRING:PUBLICATION.
[6] Z. Li, Y. Lan, dan W. Lin, “Footbridge damage detection using smartphone-recorded responses of micromobility and convolutional neural networks,” Autom. Constr., vol. 166, hlm. 105587, Okt 2024, doi: 10.1016/J.AUTCON.2024.105587.
[7] Y. Lecun, Y. Bengio, dan G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, hlm. 436–444, Mei 2015, doi: 10.1038/NATURE14539;SUBJMETA.
[8] F. Luleci dan F. N. Catbas, “A brief introductory review to deep generative models for civil structural health monitoring,” AI Civ. Eng. 2023 21, vol. 2, no. 1, hlm. 1–11, Agu 2023, doi: 10.1007/S43503-023-00017-Z.
[9] E. S. Jun, H. J. Sim, dan S. J. Moon, “Advancing YOLOv8-Based Wafer Notch-Angle Detection Using Oriented Bounding Boxes, Hyperparameter Tuning, Architecture Refinement, and Transfer Learning,” Appl. Sci. Switz., vol. 15, no. 21, Nov 2025, doi: 10.3390/app152111507.
[10] M. Bakirci, “Real-Time Vehicle Detection Using YOLOv8-Nano for Intelligent Transportation Systems,” Trait. Signal, vol. 41, no. 04, hlm. 1727–1740, 2024, doi: 10.18280/ts.410407.
[11] Y. J. Cha, R. Ali, J. Lewis, dan O. Büyükӧztürk, “Deep learning-based structural health monitoring,” Autom. Constr., vol. 161, no. February, 2024, doi: 10.1016/j.autcon.2024.105328.
[12] T. Wan dkk., “A Survey of Deep Active Learning for Foundation Models,” Intell. Comput., vol. 2, Nov 2023, doi: 10.34133/ICOMPUTING.0058;REQUESTEDJOURNAL:JOURNAL:ICOMPUTING;WEBSITE:WEBSITE:SPJ-SITE;ISSUE:ISSUE:DOI.
[13] B. D. Graham Bettis, P.E., Director, Bridge Inspections Manual, no. September. Texas, 2024.
[14] YOLO-WISLR, “Mould Dataset,” Roboflow Universe. Diakses: 27 November 2025. [Daring]. Tersedia pada: https://universe.roboflow.com/yolo-wislr/mould-0hu11
[15] Team5, “Wall Crack Dataset,” Roboflow Universe.
[16] K. Umam dan A. Ardiansyah, “Perbandingan Metode ARIMA dan LSTM pada Prediksi Jumlah Pengunjung Perpustakaan,” MIND Multimed. Artif. Intell. Netw. Database J., vol. 8, no. 2, hlm. 119–129, Des 2023, doi: 10.26760/mindjournal.v8i2.119-129.
[17] A. A. Rahman, S. D. Agustin, N. Ibrahim, dan N. C. Kumalasari, “Perbandingan Algoritma YOLOv4 dan Scaled YOLOv4 untuk Deteksi Objek pada Citra Termal,” MIND Multimed. Artif. Intell. Netw. Database J., vol. 7, no. 1, hlm. 61–71, Jun 2022, doi: 10.26760/mindjournal.v7i1.61-71.
[18] L. Fan dkk., “Impact of Image Preprocessing and Crack Type Distribution on YOLOv8-Based Road Crack Detection,” Sensors, vol. 25, no. 7, Mar 2025, doi: 10.3390/s25072180.
[19] Y. Wang dan J. He, “A Rapid Concrete Crack Detection Method Based on Improved YOLOv8,” IEEE Access, vol. 13, hlm. 59227–59243, 2025, doi: 10.1109/ACCESS.2025.3555825.
[20] M. Zand, A. Etemad, dan M. Greenspan, “Oriented Bounding Boxes for Small and Freely Rotated Objects,” IEEE Trans. Geosci. Remote Sens., vol. 60, 2022, doi: 10.1109/TGRS.2021.3076050.
[21] A. Sharma, V. Kumar, dan L. Longchamps, “Comparative performance of YOLOv8, YOLOv9, YOLOv10, YOLOv11 and Faster R-CNN models for detection of multiple weed species,” Smart Agric. Technol., vol. 9, hlm. 100648, Des 2024, doi: 10.1016/j.atech.2024.100648.
[22] R. Liu, Y. Zhu, Y. Wang, dan Z. Xing, “OB-YOLO: A UAV Image Detection Model for Reducing Computational Resource Consumption,” dalam Proceedings of the International Joint Conference on Neural Networks, Institute of Electrical and Electronics Engineers Inc., 2024. doi: 10.1109/IJCNN60899.2024.10649954.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Krisna Dzakyya Darajat, Arilyan Okta Hendriansyah
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
