請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/100200完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 張家銘 | zh_TW |
| dc.contributor.advisor | Chia-Ming Chang | en |
| dc.contributor.author | 李嘉慷 | zh_TW |
| dc.contributor.author | Lee Jia Kang | en |
| dc.date.accessioned | 2025-09-24T16:49:50Z | - |
| dc.date.available | 2025-09-25 | - |
| dc.date.copyright | 2025-09-24 | - |
| dc.date.issued | 2025 | - |
| dc.date.submitted | 2025-08-07 | - |
| dc.identifier.citation | [1] 周中哲、吳俊霖、柴駿甫、姚昭智 (2024)，2024-04-03臺灣花蓮地震事件彙整報告，國家地震工程研究中心。
[2] 內政部國土管理署, 「災害後危險建築物緊急評估辦法」. 2009. [3] Chang, P. C., Flatau, A., & Liu, S. C. (2003). Health Monitoring of Civil Infrastructure (Review). Structural Health Monitoring, 2(3), 257–267. [4] Adhikari, R. S., Moselhi, O., & Bagchi, A. (2014). Image-based retrieval of concrete crack properties for bridge inspection. Automation in Construction, 39, 180-194. [5] Cha, Y. J., Choi, W., Suh, G., Mahmoudkhani, S., & Büyüköztürk, O. (2018). Autonomous structural visual inspection using region‐based deep learning for detecting multiple damage types. Computer‐Aided Civil and Infrastructure Engineering, 33(9), 731-747. [6] Yamaguchi, T., Hashimoto, S. (2010). Fast crack detection method for large-size concrete surface images using percolation-based image processing. Machine Vision and Applications, 21, 797–809. https://doi.org/10.1007/s00138-009-0189-8 [7] Jahanshahi, M. R., Kelly, J. S., Masri, S. F., & Sukhatme, G. S. (2009). A survey and evaluation of promising approaches for automatic image-based defect detection of bridge structures. Structure and Infrastructure Engineering, 5(6), 455-486. [8] Koch, C., Georgieva, K., Kasireddy, V., Akinci, B., & Fieguth, P. (2015). A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure. Advanced engineering informatics, 29(2), 196-210 [9] Yeum, C. M., Dyke, S. J., & Ramirez, J. (2018). Visual data classification in post-event building reconnaissance. Engineering Structures, 155, 16-24. https://doi.org/10.1016/j.engstruct.2017.10.057 [10] Pan, X., & Yang, T. Y. (2020). Postdisaster imaged-based damage detection and repair cost estimation of reinforced concrete building using dual convolutional neural networks. Computer-Aided Civil and Infrastructure Engineering, 35(5), 495–510. https://doi.org/10.1111/mice.12549 [11] Ghosh Mondal, T., Jahanshahi, M. R., Wu, R. T., & Wu, Z. Y. (2020). Deep learning-based multi-class damage detection for autonomous post-disaster reconnaissance. Structural Control and Health Monitoring, 27(4), e2507. https://doi.org/10.1002/stc.2507 [12] Chiou, T. C., & Hwang, S. J. (2015). Tests on cyclic behavior of reinforced concrete frames with brick infill. Earthquake Engineering & Structural Dynamics, 44(12), 1939-1958. [13] 吳青峰（2021）。基於深度學習之RC建築震後初步耐震能力評估。﹝碩士論文。國立臺灣科技大學﹞臺灣博碩士論文知識加值系統。 [14] 姚亭君（2021）。界面滑移對於震損 RC 填充牆耐震容量之影響。 ﹝碩士論文。國立臺灣科技大學﹞臺灣博碩士論文知識加值系統。 [15] Yeh, R. L., Tseng, C. C., & Hwang, S. J. (2018). Shear strength of reinforced concrete vertical wall segments under seismic loading. ACI Structural Journal, 115(5), 1485-1494. [16] 陳建忠、葉勇凱、翁樸文、邱聰智、沈文成, 震災後危險建築物緊急評估技術及制度之修訂研究. 內政部建築研究所協同報告, 2014. [17] Chiou, T. C., Huang, C. T. & Chung, L. L. (2017). Verification on Seismic Assessment Models of Brick Infill by Tests of RC Frame with Brick infill. Journal of the Chinese Institute of Civil and Hydraulic Engineering, 29, 257-268. [18] Hassan, A. F., & Sozen, M. A. (1997). Seismic vulnerability assessment of low-rise buildings in regions with infrequent earthquakes. ACI structural journal, 94(1), 31-39. [19] Pujol, S., Laughery, L., Puranam, A., Hesam, P., Cheng, L. H., Lund, A., & Irfanoglu, A. (2020). Evaluation of seismic vulnerability indices for low-rise reinforced concrete buildings including data from the 6 February 2016 Taiwan Earthquake. Journal of Disaster Research, 15(1), 9-19. [20] 許丁友等(2003)。 「國民中小學典型校舍耐震能力初步評估法」，國家地震工程研究中心，研究報告No. NCREE-03-049，臺北。 [21] Chiou, T. C., Hwang, S. J., Chung, L. L., Tu, Y. S., Shen, W. C., & Weng, P. W. (2017). Preliminary seismic assessment of low-rise reinforced concrete buildings in Taiwan. In 16th world conference on earthquake engineering, 16WCEE. [22] 林煜衡、邱聰智、黃世建，(2018)「鋼筋混凝土住宅建築之耐震能力初步評估方法」，中華民國第十四屆結構工程研討會暨第四屆地震工程研討會，台中，論文編號：13013。 [23] 內政部國土管理署, 「建築物耐震設計規範與解說」.2011. [24] 宋嘉誠, 邱聰智, 黃世建, 臺灣中小學校舍結構耐震安全柱量比之研究. 國家地震工程研究中心, NCREE-13-031, 2013. [25] Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788). [26] Jocher, G., Chaurasia, A., & Qiu, J. (2023). YOLO by Ultralytics (Version 8.0.0) [Computer software]. https://github.com/ultralytics/ultralytics [27] Jocher, G., & Qiu, J. (2024). Ultralytics YOLOv11 documentation. Ultralytics. https://docs.ultralytics.com/models/yolo11/ [28] Chattopadhay, A., Sarkar, A., Howlader, P., & Balasubramanian, V. N. (2018). Grad-cam++: Generalized gradient-based visual explanations for deep convolutional networks. In 2018 IEEE Winter conference on Applications of Computer Vision (WACV) (pp. 839-847). IEEE. [29] RangeKing. (2023). YOLOv8 architecture visualization [Diagram]. GitHub. https://github.com/ultralytics/ultralytics/issues/189 [30] PyTorch, torch.nn.Conv2d — PyTorch Documentation, 2024. [Online]. Available: https://pytorch.org/docs/stable/generated/torch.nn.Conv2d.html [31] Ioffe, S., & Szegedy, C. (2015, June). Batch normalization: Accelerating deep network training by reducing internal covariate shift. In International conference on machine learning (pp. 448-456). pmlr. [32] Elfwing, S., Uchibe, E., & Doya, K. (2018). Sigmoid-weighted linear units for neural network function approximation in reinforcement learning. Neural networks, 107, 3-11. [33] Ramachandran, P., Zoph, B., & Le, Q. V. (2017). Searching for activation functions. arXiv preprint arXiv:1710.05941. [34] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778). [35] He, K., Zhang, X., Ren, S., & Sun, J. (2015). Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE transactions on pattern analysis and machine intelligence, 37(9), 1904-1916. [36] GeeksforGeeks. (2022). Apply a 2D max pooling in PyTorch. https://www.geeksforgeeks.org/apply-a-2d-max-pooling-in-pytorch/ [37] Chungwook Sim, Nick Skok, Ayhan Irfanoglu, Santiago Pujol, Mete Sozen, Cheng Song (2016), “Database of low-rise reinforced concrete buildings with earthquake damage,” https://datacenterhub.org/deedsdv/publications/view/454 [38] Jonathan Monical (2020), "Building Surveys after Earthquakes," https://datacenterhub.org/deedsdv/publications/view/137 [39] NCREE; Purdue University (2018), “Building Data of the 20160206 Meinong Earthquake in Taiwan,” https://www.ncree.org/recce/20160206/ [40] Chiou, T. C., Ho, Y. S., Weng, P. W. & Shen, W. C. (2018). “Building data of the 20180206 Hualien earthquake in Taiwan,” https://www.ncree.org/recce/20180206/ [41] Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision (pp. 618-626). [42] Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., & Torralba, A. (2016). Learning deep features for discriminative localization. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2921-2929). | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/100200 | - |
| dc.description.abstract | 台灣地震頻繁，建物受損風險高，震後快速評估對於安全判定至關重要。隨著顯卡算力提升與人工智慧影像辨識技術的進步，本研究旨在開發影像辨識模型以利用電腦視覺和深度學習技術自動檢測牆體損傷，評估建物安全性，判斷是否需補強或拆除，由此減少工程師的時間與人力成本。本研究透過收集世界各地包含台灣等多個著名地震資料，其中有Datacenterhub網站上來自DEEDS的數個地震資料、來自國家地震工程研究中心資料庫中的2016年美濃地震和2018年花蓮地震資料以及2024年0403花蓮地震資料等，篩選出地震後建物的承重牆的牆體照片。接著，依次將牆體照片按照RC牆和磚牆分類，再將RC依照內政部國土管理署（舊稱營建署）的五級損傷標準分級，將磚牆按照國家地震中心五級損傷標準分級。本研究將採用YOLO(You Only Look Once)一個熱門的電腦視覺模型作為本研究的模型架構，所採用的YOLO版本分別為YOLOv8和YOLO11的分類模型。將種類和級別都分好牆體圖像放入YOLOv8和YOLO11模型中分別進行訓練。以此開發出一個能夠根據RC牆的震後損傷特徵進行損傷等級分級的模型，以及另一個能夠根據磚牆的震後損傷特徵進行損傷等級分類的模型。除此之外，本研究也會使用交叉驗證的方式，將資料樣本切割成複數較小的資料子集，通過將不同的資料子集當初訓練集，而其他資料子集當成驗證集進行交叉驗證，重複進行，便可以交叉驗證。交叉驗證的目的是避免模型出現過擬合和選擇偏差等問題，同時可以反映出模型在各個不同資料集下的性能，從而得出模型的平均性能和最佳性能。本研究也採用Grad-CAM++的技術去對模型的識別能力進行視覺化檢驗。 | zh_TW |
| dc.description.abstract | Taiwan experiences frequent earthquakes, putting buildings at high risk of damage. Post-earthquake rapid assessments are necessary for determining safety. With advancements in GPU computing power and Artificial Intelligence (AI) image recognition technology, this study aims to develop an image recognition model using computer vision and deep learning techniques to automatically detect wall damage, assess building safety, and determine whether reinforcement or demolition is required, in order to reduce the time and labor costs for engineers. This study collects earthquake data from various well-known seismic events worldwide, including those in Taiwan, such as multiple datasets from the DEEDS collection available on the Datacenterhub website, as well as data from the National Center for Research on Earthquake Engineering (NCREE) in Taiwan, specifically the 2016 Meinong earthquake, the 2018 Hualien earthquake, and the 2024 April 3 Hualien earthquake, to systematically filter and analyze wall structure images of load-bearing walls in buildings affected by these earthquakes. Subsequently, the wall images are categorized into RC walls and brick walls, with RC walls classified based on the 5-level damage assessment standard established by the National Land Management Agency (formerly the Construction and Planning Agency) and brick walls classified according to the 5-level damage assessment standard of the NCREE. This study adopts YOLO (You Only Look Once), a popular computer vision model, as the core model architecture by employing the classification models YOLOv8 and YOLOv11. Labeled wall images with predefined categories and damage levels are input into the YOLOv8 and YOLOv11 models for separate training. Through this process, one model is developed to classify post-earthquake damage levels based on the labeled features of RC walls, while another model is designed to classify post-earthquake damage levels according to the labeled features of brick walls. Moreover, this study employs cross-validation by partitioning the dataset into multiple smaller subsets. In each iteration, different subsets are assigned as the training set while the remaining subsets serve as the validation set. By repeating this process, cross-validation is performed systematically. The purpose of employing cross-validation is to reduce the risk of overfitting and selection bias, while also enabling a comprehensive evaluation of the model performance across varying data splits. This approach facilitates the estimation of both the average performance and optimal performance of the model. This study also employs the Grad-CAM++ technique to examine the recognition capability of the model visually. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2025-09-24T16:49:50Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2025-09-24T16:49:50Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | 口試委員會審定書 #
誌謝 i 中文摘要 ii ABSTRACT iii CONTENTS v LIST OF FIGURES vii LIST OF TABLES xii Chapter 1 Introduction 1 1.1. Motivation and Background 1 1.2. Literature Review 4 1.3. Research Objective 19 Chapter 2 Methodology of Wall Damage Rating Model 23 2.1. Training Framework Overview 23 2.2. Model Selection and Introduction 25 2.3. Image Collection 36 2.4. Image Preprocessing 37 2.4.1. Wall Boundary Confirmation 41 2.4.2. Edge Preservation Cropping 42 2.4.3. Floor-Level Cropping 43 2.5. Cross-Validation 44 2.5.1. K-Fold Validation 44 2.5.2. Shuffle Split Validation 45 2.5.3. Stratified Sampling Strategy 46 2.5.4. Selected Cross-Validation Strategy in This Study 48 2.6. Training Procedures 50 2.6.1. Data Augmentation 51 2.6.2. Loss and Activation Function 51 2.6.3. Optimizer 52 2.6.4. Hyperparameter Configuration 54 2.7. Model Performance Evaluation Method 56 2.7.1. Numerical form Evaluation 56 2.7.2. Graphical Form Evaluation using Grad-CAM++ 57 Chapter 3 Result and Evaluation of the Wall Damage Rating Model 60 3.1. Brick Wall Result 60 3.2. RC Wall Result 77 3.3. Improvement 95 3.4. Study of Special Cases 96 Chapter 4 Practical application of the wall damage rating model 101 4.1. Post-Earthquake Rapid Assessment Evaluation 101 4.2. Model Prediction in Practical 107 Chapter 5 Conclusion and Future Work 112 5.1. Conclusions 112 5.2. Future Work 114 REFERENCES 116 | - |
| dc.language.iso | en | - |
| dc.subject | 影像辨識 | zh_TW |
| dc.subject | 交叉驗證 | zh_TW |
| dc.subject | 損傷分級 | zh_TW |
| dc.subject | 人工智慧 | zh_TW |
| dc.subject | 磚牆 | zh_TW |
| dc.subject | RC牆 | zh_TW |
| dc.subject | 牆體結構 | zh_TW |
| dc.subject | 電腦視覺 | zh_TW |
| dc.subject | Grad-CAM++ | zh_TW |
| dc.subject | Artificial Intelligence | en |
| dc.subject | Deep Learning | en |
| dc.subject | Grad-CAM++ | en |
| dc.subject | Computer Vision | en |
| dc.subject | Cross-Validation | en |
| dc.subject | Damage Rating | en |
| dc.subject | Brick Wall | en |
| dc.subject | RC Wall | en |
| dc.subject | Image Recognition | en |
| dc.title | 應用深度學習於建物RC牆和磚牆震後損傷程度分級 | zh_TW |
| dc.title | Applications of Deep Learning for Post-Earthquake Damage Rating of RC Walls and Brick Walls | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 113-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 張書瑋;林之謙;楊卓諺 | zh_TW |
| dc.contributor.oralexamcommittee | Shu-Wei Chang;Jacob J. Lin;Cho-Yen Yang | en |
| dc.subject.keyword | 人工智慧,影像辨識,電腦視覺,牆體結構,RC牆,磚牆,損傷分級,交叉驗證,Grad-CAM++, | zh_TW |
| dc.subject.keyword | Artificial Intelligence,Image Recognition,Computer Vision,Deep Learning,RC Wall,Brick Wall,Damage Rating,Cross-Validation,Grad-CAM++, | en |
| dc.relation.page | 120 | - |
| dc.identifier.doi | 10.6342/NTU202503543 | - |
| dc.rights.note | 未授權 | - |
| dc.date.accepted | 2025-08-08 | - |
| dc.contributor.author-college | 工學院 | - |
| dc.contributor.author-dept | 土木工程學系 | - |
| dc.date.embargo-lift | N/A | - |
| 顯示於系所單位： | 土木工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-113-2.pdf 未授權公開取用 | 6.77 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。