由於科技的日新月異下，人工智慧成為近年來得一個趨勢，各行各業
都想辦法結合人工智慧的概念，人工智慧為各個行業帶來了巨大的機會和
變革潛力，在人工智慧的發展下，加上硬體技術的進步，機器學習、深度
學習的概念也隨之提出，而深度學習中的卷積神經網路，讓影像辨識的準
確率有非常顯著的提升。
營建產業屬於傳統產業，而傳統產業目前正面臨轉型的階段，專業人
力的缺乏導致管理領域顯露窘境，如何讓工程管理方面有更進一步的提升，
改善過去透過大量管理人力進行繁雜的資料彙整及處理程序，加入自動化
的影像辨識技術，亦為多方研究所努力的方向。
本研究將探討影像辨識於各施工階段分類上的可能，將透過手持式裝
置、縮時攝影機、網路施工影片蒐集圖片，並透過深度學習中的卷積神經
網路，透過遷移式學習的方式，選擇經典的模型進行施工階段圖片的分類，
並比較各經典模型的效果，選擇一個適合施工階段圖片分類的模型的初探。

Due to the rapid changes in technology, artificial intelligence has become a
trend in recent years. All walks of life are trying to combine the concept of
artificial intelligence. Artificial intelligence has brought huge opportunities and
potential for change to various industries. With the advancement of hardware
technology, the concepts of machine learning and deep learning have also been
proposed, and the convolutional neural network in deep learning has
significantly improved the accuracy of image recognition.
The construction industry is a traditional industry, and the traditional
industry is currently facing the stage of transformation. The lack of professional
manpower has led to a dilemma in the management field. How to further improve
the engineering management and improve the complex data collection and
processing through a large number of management manpower in the past
Program, adding automatic image recognition technology, is also the direction
of the efforts of many parties.
This research will explore the possibility of image recognition in the
classification of various construction stages. Images will be collected through
hand-held devices, time-lapse cameras, and online construction videos, and will
be collected through convolutional neural networks in deep learning and transfer
learning. , select the classic model to classify the pictures in the construction
stage, and compare the effects of each classic model, and select a model suitable
for the classification of pictures in the construction stage.

[1] 王聖凱(2020)-整合擴增實境與BIM之施工階段進度間空系統
[2] 謝政廷(2021)- 整合 SLAM-based頭戴式 MR與BIM之室內施工進度視覺化管理遠端協作系統
[3] 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.
[4] Pratt L. Y.,Discriminability-based transfer between neural networks. NIPS Conference: Advances in Neural Information Processing Systems 5. Morgan Kaufmann Publishers. 1993: 204–211.
[5] Langley P. (1986). Human and machine learning. Machine Learning, Volume 1, Pages 243–248.
[6] 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
[7] Ferreiro-Cabello J., Fraile-Garcia E., Martinez de Pison Ascacibar E., Martinez de Pison Ascacibar F.J. (2018). Metamodel-based design optimization of structural one-way slabs based on deep learning neural networks to reduce environmental impact. Engineering Structures, Volume 155, Pages 91-101.
[8] Chen J., Kira Z., Cho Y.K. (2019). Deep Learning Approach to Point Cloud Scene Understanding for Automated Scan to 3D Reconstruction. Journal of Computing in Civil Engineering, Volume 33, Issue 4.
[9] Bunrit S., Kerdprasop N., Kerdprasop K. (2019). Evaluating on the transfer learning of CNN architectures to a construction material image classification tasks. International Journal of Machine Learning and Computing, Volume 9, Issue 2, Pages 201-207.
[10] 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.
[11] 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.
[12] 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
[13] Aravinda S. R., Nguyen T., Palaniswami M., Ngo T.(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
[14] 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
[15] 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
[16] ImageNet,[online], https://image-net.org/
[17] Gao Y., Mosalam K. M.(2018),Deep Transfer Learning for Image-Based Structural Damage Recognition. Computer-Aided Civil and Infrastructure Engineering, Volume 33 Issue9,Special Issue: Health Monitoring of Structures, Pages 748–768
[18] 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
[19] Bunrit S., Kerdprasop N., Kerdprasop K.(2019), Evaluating on the Transfer Learning of CNN Architectures to a Construction Material Image Classification Task. International Journal of Machine Learning and Computing, Volume 9,Pages 201-207
[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] Python,[online],https://www.python.org/
[22] Tensorflow,[online],https://www.tensorflow.org/
[23] Keras,[online], https://keras.io/
[24] scikit-learn,[online],https://scikit-learn.org/stable/
[25] Simonyan K., Zisserman A., Very Deep Convolutional Networks For Large-Scale Image Recognition, arXiv:1409.1556v6
[26] Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Erhan D., Vanhoucke V., Rabinovich A., Going deeper with convolutions, arXiv:1409.4842v1
[27] Zhang X.,He K., Ren S., Sun J., Deep Residual Learning for Image Recognition, arXiv:1512.03385v1
[28] Huang G., Liu Z., Maaten L. V. D., Weinberger K. Q., Densely Connected Convolutional Networks, arXiv:1608.06993v5
[29] Chollet o., Franc,Xception: Deep Learning with Depthwise Separable Convolutions, arXiv:1610.02357v3
[30] Szegedy C., Ioffe S., Vanhoucke V., Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning, arXiv:1602.07261v2
[31] Howard A. G., Zhu M., Chen B., Kalenichenko D., Wang W., Weyand T., Andreetto M., Adam H., MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications, arXiv:1704.04861v1
[32] Tan M., Le Q. V., EfficientNetV2: Smaller Models and Faster Training, arXiv:2104.00298v3
[33] Van Rijsbergen, C. J. (1979). Information Retrieval (2nd ed.). London: Butterworths.
[34] Fawcett, Tom (2006); An introduction to ROC analysis, Pattern Recognition Letters, 27, 861–874.