近年來隨著科技進步，人工智慧的影像辨識技術已發展相當成熟，並已廣泛運用於各項產業，建築資訊模型(Building Information Model, BIM)的建構與應用近年來已成為營建產業實務上數位化的主流技術，本研究提出結合BIM模型與影像辨識技術使工地施工進度管控達到自動化。
為達該目的使用到影像辨識中的物件偵測技術，並建立施工進度數據集，也運用深度學習的技術，嘗試不同卷積神經網路與物件偵測的模型組合，找出最合適之模型組合，並透過遷移式學習的方式，優化此模型所選用的函數、權重與參數，經過適當之訓練、測試、與驗證後針對本研究之應用提出AI影像偵測準確率最佳的模型。
另於AI影像偵測結果與BIM模型整合方面，將攝影機擷取的影像與對應視角的模型畫面套合後，透過二者間的座標轉換，即可將各個影像上的偵測結果輸入至應用端與BIM模型進行整合，取得該構件之施工進度。

The recent advancements in technology have led to the maturity of artificial intelligence image recognition techniques, which have been widely employed across various industries. The establishment and implementation of Building Information Models has become a widely accepted digitalization trend within the construction industry. This research endeavors to integrate BIM with image recognition technology to realize an automated system for construction site progress monitoring and control.
The objective is attained by utilizing object detection techniques from the field of image recognition and constructing a dataset for construction progress. Deep learning techniques are also employed to experiment with various combinations of convolutional neural networks and object detection models, with the aim of identifying the optimal combination. The chosen model is optimized through transfer learning, which involves adjusting the functions, weights, and parameters used. After appropriate training, testing, and validation, the model with the highest accuracy for AI image detection is proposed for implementation in this research.
With regards to integrating the results of AI image detection with the BIM model, the captured image from the camera is aligned with the corresponding view of the model. Through coordinate transformation between the two, the detection results in each image can be input into the application and integrated with the BIM model, thus providing information on the construction progress of the component.

[1] 陳宇翔(2022),基於卷積神經網路之施工進度圖片 AI 分類,國立臺灣科技大學碩士學位論文
[2] McCarthy J., Minsky M. L., Rochester N., Shannon C.E. (2006). A proposal for the Dartmouth summer research project on artificial intelligence. AI Magazine, Volume 27, Issue 4, Pages 12-14.
[3] Langley P. (1986). Human and machine learning. Machine Learning, Volume 1, Pages 243–248.
[4] Hinton G. E., Salakhutdinov R. R. , Reducing the Dimensionality of Data with Neural Networks, Science,28 Jul 2006,Vol 313, Issue 5786,pp. 504-507
[5] Krizhevsky A., Sutskever I., Hinton G. E. (2012). ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, Volume 2, Pages 1097-1105.
[6] Albawi S., Mohammed T. A., Al-Zawi S. (2017). Understanding of a convolutional neural network. International Conference on Engineering and Technology (ICET).
[7] 林子皓(2020),以圖像分析方法辨識潛在異常施工活動, 國立臺灣大學碩士學位論文
[8] 鄭群嚴(2020),工地安全之跌倒偵測－以深度學習為工具, 國立臺灣大學碩士學位論文,2020.
[9] 呂雅涵(2020),大數據深度學習技術於結構物防災監測之應用研究, 中國科技大學碩士學位論文
[10] Saudi M. M. , Ma’arof A. H. , Ahmad A. , Saudi A. S. M., Ali M. H., Narzullaev A. , Ghazali M. I. M.(2020),Image Detection Model for Construction Worker Safety Conditions using Faster R-CNN. IJACSA, Volume 11
[11] Son H., Choi H., Seong H.,Kim C.(2019), Detection of construction workers under varying poses and changing background in image sequences via very deep residual networks, Automation in Construction, Volume 99, Pages 27-38
[12] Arabi S., Haghighat A., Sharma A. (2020). A deep-learning-based computer vision solution for construction vehicle detection. Computer-Aided Civil and Infrastructure Engineering, Volume 35, Issue 7, Pages 753-767.
[13] Alipour M., Harris D. K. (2020). Increasing the robustness of material-specific deep learning models for crack detection across different materials. Engineering Structures, Volume 206.
[14] 彭大倫(2020),基於鋼構廠生產線模擬生成資料之大數據分析研究,國立臺灣科技大學碩士學位論文
[15] Aravinda S. Rao, Tuan Nguyen, Marimuthu Palaniswami, Tuan Ngo(2021), Vision-based Automated Crack Detection using Convolutional Neural Networks for Condition Assessment of Infrastructure, Structural Health Monitoring, Volume: 20 issue: 4, page(s): 2124-2142
[16] Zhang Z., Xue J., Zhang J., Yang M., Meng B., Tan Y., Ren S.(2021) A deep learning automatic classification method for clogging pervious pavement. Construction and Building Materials, Volume 309, 125195
[17] Dais D., Bal I. E., Smyrou E., Sarhosis V.(2021), Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning. Automation in Construction, Volume 125, 103606
[18] MS COCO[online], https://cocodataset.org/
[19] Kim H., Kim H., Hong Y. W., Byun H.(2018), Detecting Construction Equipment Using a Region-Based Fully Convolutional Network and Transfer Learning. Journal of Computing in Civil Engineering, Volume 32, Issue 2
[20] Yang X., Zhang Y., Lv W., Wang D.(2021),Image recognition of wind turbine blade damage based on a deep learning model with transfer learning and an ensemble learning classifier. Renewable Energy, Volume 163, January 2021, Pages 386-397
[21] 丁文佑(2022),利用遷移學習與物件偵測建立人像多屬性辨識系統, 國立高雄科技大學碩士學位論文
[22] Python,[online],https://www.python.org/
[23] Tensorflow,[online],https://www.tensorflow.org/
[24] Pytorch [online],https://www.pytorch.org/
[25] Unity,[online], https://unity.com/
[26] Everingham, M., Eslami, S. M. A., Van Gool, L., Williams, C. K. I., Winn, J. and Zisserman(2005), The PASCAL Visual Object Classes Challenge: A Retrospective A.International Journal of Computer Vision, 111(1), 98-136, 2015
[27] Redmon J., Divvala S., Girshick R., Farhadi A.(2016), You Only Look Once: Unified, Real-Time Object Detection, IEEE Computer Vision and Pattern Recognition, pages. 779-788.
[28] Yan, Bin, et al. (2021) ,A real-time apple targets detection method for picking robot based on improved YOLOv5., Remote Sensing 13.9: 1619.
[29] YOLOv5,[online]https://github.com/ultralytics/yolov5
[30] Huang J., Rathod V., Sun C., Zhu M., Korattikara A. Fathi A., Fischer I., Wojna Z., Song Y., Guadarrama S., Murphy K.(2017), Google Research,Speed/accuracy trade-offs for modern convolutional object detectors.
[31] Zhang X.,He K., Ren S., Sun J., Deep Residual Learning for Image Recognition, arXiv:1512.03385v1
[32] Huang G., Liu Z., Maaten L. V. D., Weinberger K. Q., Densely Connected Convolutional Networks, arXiv:1608.06993v5
[33] Bochkovskiy A., Chien-Yao Wang, Hong-Yuan Mark Liao.(2020), YOLOv4: Optimal Speed and Accuracy of Object Detection, (cs.CV), 23 Apr,
[34] Ren S., He K., Girshick R., Sun J.(2015), Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks, arXiv:1506.01497
[35] Zhang H., Cisse M., Dauphin Y.N.,Lopez-Paz D.(2017), mixup: Beyond Empirical Risk Minimization, arXiv:1710.09412
[36] Yun S., Han D., Oh S.J., Chun S., Choe J., Yoo Y.(2019), CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features, arXiv:1905.04899v2
[37] Hendrycks D., Mu N., Cubuk E. D., Zoph B., Gilmer J., Lakshminarayanan B.(2020), AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty, arXiv:1912.02781
[38] Cubuk E. D., Zoph B., Mané D., Vasudevan V., Le Q.V. (2019), AutoAugment: Learning Augmentation Strategies From Data
[39] Pascal VOC [online], http://host.robots.ox.ac.uk/pascal/VOC/