Computer Vision Metodlarıyla Çeşitli Mimari Üslupların Tahmin Edilmesi

Yapay zeka (AI) alanının alt dalı olan Computer Vision (bilgisayar görüşü, CV), bilgisayarların görsel verileri işleyerek nesneleri tanıyabilmesine olanak sağlar. CV, otomotiv, gıda endüstrisi, hastalıkların teşhisi gibi alanlarda yaygın kullanılmaktadır. AI bunu yaparken, algoritmaları kullanmaktadır. Nesne algılamaya dayalı algoritmaların en önemlilerinden biri yüksek veri işleme hızıyla daha net sonuçlar veren YOLO (You Only Look Once) dur. Bu çalışmanın amacı, temel alınan videodaki öne çıkan yapıların gotik, barok, palladyen, art nouveau mimari üsluplarından hangisine ait olduğunu belirlemeye yönelik nesne algılama tabanlı CV projesi gerçekleştirmektir. Çalışma veri seti oluşturma, veri etiketleme, model oluşturma ve modelin eğitimi aşamalarından oluşmaktadır. Veri etiketleme platformu olarak Roboflow, model oluşturma ve eğitim aşamaları için YOLOv8 kullanılmıştır. Süreç sonunda modelin mimari üslupları yüksek doğruluk payı ile kısa zamanda tahmin etmesi modelin başarılı gerçek zamanlı bir nesne algılama algoritması olduğunu ortaya koymuş, CV’ın mimarlık alanında da kullanılabileceği ve mimarlık ile ilgili diğer alanlara da katkı sunabileceği vurgulanmıştır.

Anahtar Kelimeler:

Computer vision, nesne algılama, YOLO, mimari üslup

Predicting Various Architectural Styles Using Computer Vision Methods

Computer Vision (CV), subfield of artificial intelligence (AI), enables computers to process visual data and recognize objects. CV is widely used in, automotive, food industry and diseases diagnosis. AI achieves this by algorithms. One of the important algorithms based on object detection is YOLO (You Only Look Once), provides more accurate results with high processing speed. The aim of this study is to perform an object detection-based CV project, to determine the structures in given video belong to one of the architectural styles: Gothic, Baroque, Palladian, or Art Nouveau. The study consists of data set creation, data labeling, model creation and model training. Roboflow was used as the data labeling platform and YOLOv8 was used for model building and training phases. At the end of the process, the fact that the model predicts architectural styles with high accuracy in a short time revealed that the model is a successful real-time object detection algorithm, and it was emphasized that CV can be used in the field of architecture and can contribute to other fields related to architecture.

Keywords:

Computer vision, object detection, YOLO, architectural style,

PDF

___

Boesch, G. (2023a, January 20). What is Computer Vision? The Complete Tech Guide for 2023 - viso.ai. Viso.Ai. http://viso.ai/computer-vision/what-is-computer-vision/.
Boesch, G. (2023b, February 21). Object Detection in 2023: The Definitive Guide - viso.ai. Viso.Ai. http://viso.ai/deep-learning/object-detection/
Contributors to Wikimedia projects. (2001a, August 8). Casa Batlló - Wikipedia. Retrieved September 25, 2023, from Wikipedia, the free encyclopedia website: https://en.wikipedia.org/wiki/Casa_Batllo
Contributors to Wikimedia projects. (2001b, October 26). Architecture - Wikipedia. Retrieved March 2, 2023, from Wikipedia, the free encyclopedia website: http://en.wikipedia.org/wiki/Architecture
Contributors to Wikimedia projects. (2002a, May 24). Palace of Versailles - Wikipedia. Retrieved September 25, 2023, from Wikipedia, the free encyclopedia website: https://en.wikipedia.org/wiki/Palace_of_Versailles
Contributors to Wikimedia projects. (2002b, July 15). Notre-Dame de Paris - Wikipedia. Retrieved September 25, 2023, from Wikipedia, the free encyclopedia website: https://en.wikipedia.org/wiki/Notre-Dame_de_Paris
Contributors to Wikimedia projects. (2003a, May 6). Milan Cathedral - Wikipedia. Retrieved September 25, 2023, from Wikipedia, the free encyclopedia website: https://en.wikipedia.org/wiki/Milan_Cathedral
Contributors to Wikimedia projects. (2003b, October 6). Trevi Fountain - Wikipedia. Retrieved September 25, 2023, from Wikipedia, the free encyclopedia website: https://en.wikipedia.org/wiki/Trevi_Fountain
Contributors to Wikimedia projects. (2006, May 17). Palladian villas of the Veneto - Wikipedia. Retrieved September 25, 2023, from Wikipedia, the free encyclopedia website: https://en.wikipedia.org/wiki/Palladian_villas_of_the_Veneto
Contributors to Wikimedia projects. (2008, February 18). Object detection - Wikipedia. Retrieved March 10, 2023, from Wikipedia, the free encyclopedia website: http://en.wikipedia.org/wiki/Object_detection
Dwyer, B. (2020, May 8). When Should I Auto-Orient My Images? Roboflow Blog; Roboflow Blog. http://blog.roboflow.com/exif-auto-orientation/
Efe, M. O. & Kaynak, O. (1999). A comparative study of neural network structures in identification of nonlinear systems. Mechatronics, 3, 287–300. https://doi.org/10.1016/s0957-4158(98)00047-6
Elmas, Ç. (2018). Yapay Zeka Uygulamaları (4th ed.). Seçkin.
Handuo. (2018, August 20). You only look once (YOLO) -- (1) | Zhang Handuo’s Site. Zhang Handuo’s Site; Zhang Handuo’s Site. http://zhanghanduo.github.io/post/yolo1/.
Hosni, Y. (2022, October 14). Overview of Computer Vision Tasks & Applications. Pub.Towardsai.Net; Towards AI. https://pub.towardsai.net/overview-of-the-computer-vision-tasks-applications-647f63e66e9f
Jocher, G. & Waxmann, S. (2023, May 1). YOLOv8 - Ultralytics YOLOv8 Docs. Ultralytics. https://docs.ultralytics.com/models/yolov8/
Jocher, G., Waxmann, S. & Chaurasia, A. (2023, March 12). Ultralytics YOLOv8 Modes. Ultralytics YOLOv8 Docs. http://docs.ultralytics.com/#yolo-a-brief-history.
Kasper-Eulaers, M., Hahn, N., Berger, S., Sebulonsen, T., Myrland, Ø. & Kummervold, P. E. (2021). Short communication: detecting heavy goods vehicles in rest areas in winter conditions using YOLOv5. Algorithms, 4, 114. https://doi.org/10.3390/a14040114
Kristo, M., Ivasic-Kos, M. & Pobar, M. (2020). Thermal Object Detection in Difficult Weather Conditions Using YOLO. IEEE Access, 125459–125476. https://doi.org/10.1109/access.2020.3007481
Özel, M. A., Baysal, S. S. & Şahin, M. (2021). Derin öğrenme algoritması (YOLO) ile dinamik test süresince süspansiyon parçalarında çatlak tespiti. Avrupa Bilim ve Teknoloji Dergisi, Ejosat, 1–5. https://doi.org/10.31590/ejosat.952798
Öztürkoğlu, M. (2023a, April 25). Architectural Buildings3 Computer Vision Project. Roboflow. http://app.roboflow.com/meryem-dgz60/architecturalbuildings3/1
Öztürkoğlu, M. (2023b, April 30). Estimating Various Architectural Styles with Computer Vision Methods. Google Colab. https://colab.research.google.com/drive/1ldJ4P2tMJhCaK7j7LxO-ct3UygW9ERCq?usp=sharing
Öztürkoğlu, M. (2023c, May 9). Before Train_Estimating Various Architectural Styles with Computer Vision Methods. Youtube.Com; YouTube. https://www.youtube.com/watch?v=bgctNx_1luE
Öztürkoğlu, M. (2023d, May 10). Estimating Various Architectural Styles with Computer Vision Methods. Youtube.Com; YouTube. https://www.youtube.com/watch?v=CC7fakCsCSM
Rath, S. (2023, January 10). YOLOv8 Ultralytics: State-of-the-Art YOLO Models. LearnOpenCV. http://learnopencv.com/ultralytics-yolov8/.
Redmon, J., Divvala, S., Girshick, R. & Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). https://doi.org/10.1109/cvpr.2016.91
Roboflow. (2020, January). Roboflow. https://roboflow.com
Sager, C., Janiesch, C. & Zschech, P. (2021). A survey of image labelling for computer vision applications. Journal of Business Analytics,2,91–110.https://doi.org/10.1080/2573234x.2021.1908861
Simplilearn. (2022, August 30). What is Epoch in Machine Learning? | Simplilearn. Retrieved September 25, 2023, from Simplilearn.com website: https://www.simplilearn.com/tutorials/machine-learning-tutorial/what-is- epoch-in-machine-learning?tag=epoch
Su, C. (2008, April 5). Introduction to Computer Vision. Carleton.Ca; National Research Council Canada. https://people.scs.carleton.ca/~c_shu/Courses/comp4900d/notes/lect1_intro.pdf
Szeliski, R. (2010). Computer Vision (1st ed., p. 5). Springer.
Terven, J. & Cordova-Esparza, Diana-Margarita. (2023). A Comprehensive Review of YOLO: From YOLOv1 to YOLOv8 and Beyond.
Trucco, E. & Verri, A. (1998). Introductory Techniques for 3-D Computer Vision. Prentice Hall.
Vitruvius. (1999). Vitruvius: “Ten Books on Architecture” (I. D. Rowland, Ed.; T. N. Howe, Trans.). Cambridge University Press.
Williams, K. (2021, July 5). How to Build a Computer Vision Model. Medium. http://medium.com/mlearning- ai/what-does-end-to-end-really-mean-f634b193ba00.
Wwymak. (n.d.). Architecture dataset | Kaggle. Kaggle: Your Machine Learning and Data Science Community. Retrieved June 28, 2023, from http://www.kaggle.com/datasets/wwymak/architecture-dataset
Xu, Z., Tao, D., Zhang, Y., Wu, J., & Tsoi, A. C. (2014). Architectural Style Classification Using Multinomial Latent Logistic Regression. In Computer Vision – ECCV 2014 (pp. 600–615). Springer International Publishing. http://dx.doi.org/10.1007/978-3-319-10590-1_39
Yıldız, M. A., Ertosun Yıldız, M. & Beyhan, F. (2023). Developing dynamic and flexible façade design with fractal geometry. Journal of Architectural Sciences and Applications, 8 (1), 1-14. DOI: 10.30785/mbud.1230875.