Derin öğrenme uygulamalarında kullanılan uzaktan algılama verilerinden oluşturulmuş açık kaynaklı bina veri setleri: Karşılaştırmalı değerlendirme

Yıl 2024, , 1 - 11, 15.04.2024

Esra Özaydın , Burcu Amirgan , Gülşen Taşkın , Nebiye Musaoğlu

https://doi.org/10.29128/geomatik.1257555

Öz

Bina çıkarımı; arazi kullanımı, şehir planlaması, afet izleme, navigasyon, coğrafi veri tabanlarının güncellenmesi ve kentsel dinamik izleme gibi çeşitli mekânsal uygulamalarda önemli rol oynar. Farklı bölgelerdeki binalar farklı yapısal ve geometrik özelliklere sahip olduğundan görüntülerden otomatik bina çıkarımı zor bir iştir. Son yıllarda uygun veri setleriyle eğitildiklerinde klasik makine öğrenme yöntemlerine göre daha yüksek doğruluklu sonuçlar üreten derin öğrenme modelleri, otomatik bina çıkarımında sıkça kullanılmaktadır. Modellerin yüksek doğrulukta eğitilmesi için kaliteli etiketlerin olduğu bina veri setleri büyük önem taşımaktadır. Bu çalışmanın amacı, bina tespiti için farklı çözünürlükteki uzaktan algılama görüntülerinden oluşturulmuş ve literatürde sıkça kullanılan açık kaynaklı bina veri setlerini tanıtmaktır. Veri setleri, kaydedildiği platformlara göre havadan, uydudan ve her iki platformdan kaydedilmiş görüntülerden oluşan veriler olarak üç kategoride gruplandırılıp, detayları açıklanmıştır. Bunun yanı sıra veri setleri ile yapılmış karşılaştırmalı çalışmaları içeren güncel literatür özeti verilmiştir. Bina tespiti işlemini doğru şekilde gerçekleştirmek için araştırmacılara rehberlik edecek ve bina veri seti oluşturulmasında dikkat edilmesi gereken kritik hususları içeren değerlendirmeler sunulmuştur.

Anahtar Kelimeler

Bina Çıkarımı, Bina Veri Setleri, Derin Öğrenme, Uzaktan Algılama

Kaynakça

• Akbulut, Z., Özdemir, S., Acar, H., Dihkan, M., & Karslı, F. (2018). Automatic extraction of building boundaries from high resolution images with active contour segmentation. International Journal of Engineering and Geosciences, 3(1), 36-42. https://doi.org/10.26833/ijeg.373152
• Amirgan, B., Awad, B., Erer, I., & Musaoğlu, N. (2022). A comparative study for building segmentation in remote sensing images using deep networks: Cscrs Istanbul building dataset and results. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 46, 1-6. https://doi.org/10.5194/isprs-archives-XLVI-M-2-2022-1-2022
• Atik, M. E., Duran, Z., & Özgünlük, R. (2022). Comparison of YOLO versions for object detection from aerial images. International Journal of Environment and Geoinformatics, 9(2), 87-93. https://doi.org/10.30897/ijegeo.1010741
• Azam, B., Khan, M. J., Bhatti, F. A., Maud, A. R. M., Hussain, S. F., Hashmi, A. J., & Khurshid, K. (2022). Aircraft detection in satellite imagery using deep learning-based object detectors. Microprocessors and Microsystems, 94, 104630. https://doi.org/10.1016/j.micpro.2022.104630
• Bakirman, T., Komurcu, I., & Sertel, E. (2022). Comparative analysis of deep learning based building extraction methods with the new VHR Istanbul dataset. Expert Systems with Applications, 202, 117346. https://doi.org/10.1016/j.eswa.2022.117346
• Bayramoğlu, Z., & Uzar, M. (2023). Performance analysis of rule-based classification and deep learning method for automatic road extraction. International Journal of Engineering and Geosciences, 8(1), 83-97. https://doi.org/10.26833/ijeg.1062250
• Boguszewski, A., Batorski, D., Ziemba-Jankowska, N., Dziedzic, T., & Zambrzycka, A. (2021). LandCover. ai: Dataset for automatic mapping of buildings, woodlands, water and roads from aerial imagery. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1102-1110.
• Chen, Q., Wang, L., Wu, Y., Wu, G., Guo, Z., & Waslander, S. L. (2019). Temporary removal: Aerial imagery for roof segmentation: A large-scale dataset towards automatic mapping of buildings. ISPRS Journal of Photogrammetry and Remote Sensing, 147, 42-55. https://doi.org/10.1016/j.isprsjprs.2018.11.011
• Duman, H. S., & Başaraner, M. (2022). Şekil göstergeleri ve topluluk öğrenmesi sınıflandırma algoritmaları ile bina detaylarının şekil karmaşıklık analizi. Geomatik, 7(3), 197-208. https://doi.org/10.29128/geomatik.947334
• Erdem, F., & Avdan, U. (2020). Comparison of different U-net models for building extraction from high-resolution aerial imagery. International Journal of Environment and Geoinformatics, 7(3), 221-227. https://doi.org/10.30897/ijegeo.684951
• Gerke, M., Rottensteiner, F., Wegner, J., Sohn, G., 2014. ISPRS Semantic Labeling Contest. https://doi.org/10.13140/2.1.3570.9445
• Open Cities AI Challenge Dataset. Version 1.0, (2023). Radiant MLHub. https://mlhub.earth/10.34911/rdnt.f94cxb
• Glinka, S., Owerko, T., & Tomaszkiewicz, K. (2022). Using open vector-based spatial data to create semantic datasets for building segmentation for raster data. Remote Sensing, 14(12), 2745. https://doi.org/10.3390/rs14122745
• Gupta, R., Goodman, B., Patel, N., Hosfelt, R., Sajeev, S., Heim, E., ... & Gaston, M. (2019a). Creating xBD: A dataset for assessing building damage from satellite imagery. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, 10-17.
• Gupta, R., Hosfelt, R., Sajeev, S., Patel, N., Goodman, B., Doshi, J., ... & Gaston, M. (2019b). xbd: A dataset for assessing building damage from satellite imagery. Computer Vision and Pattern Recognition. https://doi.org/10.48550/arXiv.1911.09296
• He, H., Gao, K., Tan, W., Wang, L., Chen, N., Ma, L., & Li, J. (2022b). Super-resolving and composing building dataset using a momentum spatial-channel attention residual feature aggregation network. International Journal of Applied Earth Observation and Geoinformation, 111, 102826. https://doi.org/10.1016/j.jag.2022.102826
• He, H., Jiang, Z., Tan, W., Cai, Y., Fatholahi, S. N., Gao, K., ... & Li, J. (2021). Waterloo Building Dataset: A large-scale very-high-spatial-resolution image dataset for building rooftop extraction. Abstracts of the ICA, 3, 1-2. https://doi.org/10.5194/ica-abs-3-105-2021
• He, H., Jiang, Z., Gao, K., Narges Fatholahi, S., Tan, W., Hu, B., ... & Li, J. (2022a). Waterloo building dataset: A city-scale vector building dataset for mapping building footprints using aerial orthoimagery. Geomatica, 75(3), 99-115. https://doi.org/10.1139/geomat-2021-0006
• Hinton, G. E., Osindero, S., & Teh, Y. W. (2006). A fast learning algorithm for deep belief nets. Neural Computation, 18(7), 1527-1554. https://doi.org/10.1162/neco.2006.18.7.1527
• Ji, S., Wei, S., & Lu, M. (2018). Fully convolutional networks for multisource building extraction from an open aerial and satellite imagery data set. IEEE Transactions on Geoscience and Remote Sensing, 57(1), 574-586. https://doi.org/10.1109/TGRS.2018.2858817
• Jiwani, A., Ganguly, S., Ding, C., Zhou, N., & Chan, D. M. (2021). A semantic segmentation network for urban-scale building footprint extraction using RGB satellite imagery. Computer Vision and Pattern Recognition https://doi.org/10.48550/arXiv.2104.01263
• Karhunen, J., Raiko, T., & Cho, K. (2015). Unsupervised deep learning: A short review. Advances in Independent Component Analysis and Learning Machines, 125-142. https://doi.org/10.1016/B978-0-12-802806-3.00007-5
• Luo, M., Ji, S., & Wei, S. (2023). A diverse large-scale building dataset and a novel plug-and-play domain generalization method for building extraction. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 16, 4122-4138. https://doi.org/10.1109/JSTARS.2023.3268176
• Maggiori, E., Tarabalka, Y., Charpiat, G., & Alliez, P. (2017). Can semantic labeling methods generalize to any city? The Inria aerial image labeling benchmark. 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 3226-3229. 10.1109/IGARSS.2017.8127684
• Mehran, A., Tehsin, S., & Hamza, M. (2023). An effective deep learning model for ship detection from satellite images. Spatial Information Research, 31(1), 61-72. https://doi.org/10.1007/s41324-022-00482-1
• Mnih, V. (2013). Machine learning for aerial image labeling. University of Toronto (Canada).
• Mohanty, S. P., Czakon, J., Kaczmarek, K. A., Pyskir, A., Tarasiewicz, P., Kunwar, S., ... & Schilling, M. (2020). Deep learning for understanding satellite imagery: An experimental survey. Frontiers in Artificial Intelligence, 3, 534696. https://doi.org/10.3389/frai.2020.534696
• URL-1: https://haberler.itu.edu.tr/docs/default-source/default-document-library/2023_itu_deprem_on_raporu.pdf?sfvrsn=bf82d8e5_
• URL-2: https://www.isprs.org/education/benchmarks/UrbanSemLab/semantic-labeling.aspx
• URL-3: https://www.cs.toronto.edu/~vmnih/data/
• URL-4: https://project.inria.fr/aerialimagelabeling/
• URL-5: https://competitions.codalab.org/competitions/20100
• URL-6: https://www.kaggle.com/datasets/adrianboguszewski/landcoverai
• URL-7: https://www.kaggle.com/datasets/atilol/aerialimageryforroofsegmentation
• URL-8: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/EXRA2V
• URL-9: https://github.com/sajmonogy/keras_segmentation_models
• URL-10: https://spacenet.ai/spacenet-buildings-dataset-v1/
• URL-11:https://spacenet.ai/spacenet-buildings-dataset-v2/
• URL-12: https://www.aicrowd.com/challenges/mapping-challenge
• URL-13: https://spacenet.ai/off-nadir-building-detection/
• URL-14: https://xview2.org/dataset
• URL-15: https://spacenet.ai/sn6-challenge/
• URL-16: https://spacenet.ai/sn7-challenge/
• URL-17: http://rs.ipb.uni-bonn.de/data/semcity-toulouse-data-access/
• URL-18: https://sites.research.google/open-buildings/#download
• URL-19: http://gpcv.whu.edu.cn/data/building_dataset.html
• URL-20: http://gpcv.whu.edu.cn/data/whu-mix(raster)/whu_mix%20(raster).html
• URL-21: https://www.geoportal.gov.pl/
• Ozturk, O., Saritürk, B., & Seker, D. Z. (2020). Comparison of fully convolutional networks (FCN) and U-Net for road segmentation from high resolution imageries. International Journal of Environment and Geoinformatics, 7(3), 272-279. https://doi.org/10.30897/ijegeo.737993
• Patel, K., Bhatt, C., & Mazzeo, P. L. (2022). Deep learning-based automatic detection of ships: An experimental study using satellite images. Journal of Imaging, 8(7), 182. https://doi.org/10.3390/jimaging8070182
• Perihanoğlu, G. M., Özerman, U., & Şeker, D. Z. (2018). Kenar algılama ve morfoloji operatörleri kullanılarak detay çıkarımı üzerine bir uygulama. Geomatik, 3(2), 120-128. https://doi.org/10.29128/geomatik.358957
• Ps, P., & Aithal, B. H. (2023). Building footprint extraction from very high-resolution satellite images using deep learning. Journal of Spatial Science, 68(3), 487-503. https://doi.org/10.1080/14498596.2022.2037473
• Raghavan, R., Verma, D. C., Pandey, D., Anand, R., Pandey, B. K., & Singh, H. (2022). Optimized building extraction from high-resolution satellite imagery using deep learning. Multimedia Tools and Applications, 81(29), 42309-42323. https://doi.org/10.1007/s11042-022-13493-9
• Roscher, R., Volpi, M., Mallet, C., Drees, L., & Wegner, J. D. (2020). SemCity Toulouse: A benchmark for building instance segmentation in satellite images. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 5, 109-116. https://doi.org/10.5194/isprs-annals-V-5-2020-109-2020
• Sarker, I. H. (2021). Deep learning: a comprehensive overview on techniques, taxonomy, applications and research directions. SN Computer Science, 2(6), 420. https://doi.org/10.1007/s42979-021-00815-1
• Saritürk, B., Bayram, B., Duran, Z., & Seker, D. Z. (2020). Feature extraction from satellite images using segnet and fully convolutional networks (FCN). International Journal of Engineering and Geosciences, 5(3), 138-143. https://doi.org/10.26833/ijeg.645426
• Shermeyer, J., Hogan, D., Brown, J., Van Etten, A., Weir, N., Pacifici, F., ... & Lewis, R. (2020). SpaceNet 6: Multi-sensor all weather mapping dataset. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition workshops, 196-197.
• Sirko, W., Kashubin, S., Ritter, M., Annkah, A., Bouchareb, Y. S. E., Dauphin, Y., ... & Quinn, J. (2021). Continental-scale building detection from high resolution satellite imagery. Computer Vision and Pattern Recognition. https://doi.org/10.48550/arXiv.2107.12283
• Tan, C., Sun, F., Kong, T., Zhang, W., Yang, C., & Liu, C. (2018). A survey on deep transfer learning. Artificial Neural Networks and Machine Learning–ICANN 2018: 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4-7, 2018, Proceedings, Part III 27, 270-279. https://doi.org/10.1007/978-3-030-01424-7_27
• Van Etten, A., Hogan, D., Manso, J. M., Shermeyer, J., Weir, N., & Lewis, R. (2021). The multi-temporal urban development spacenet dataset. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 6398-6407.
• Van Etten, A., Lindenbaum, D., & Bacastow, T. M. (2018). Spacenet: A remote sensing dataset and challenge series. Computer Vision and Pattern Recognition. https://doi.org/10.48550/arXiv.1807.01232
• Wang, X., Liu, Y., & Xin, H. (2021). Bond strength prediction of concrete-encased steel structures using hybrid machine learning method. Structures, 32, 2279-2292. https://doi.org/10.1016/j.istruc.2021.04.018
• Weir, N., Lindenbaum, D., Bastidas, A., Etten, A. V., McPherson, S., Shermeyer, J., ... & Tang, H. (2019). Spacenet mvoi: A multi-view overhead imagery dataset. Proceedings of the IEEE/CVF International Conference on Computer Vision, 992-1001.
• Yin, J., Wu, F., Qiu, Y., Li, A., Liu, C., & Gong, X. (2022). A multiscale and multitask deep learning framework for automatic building extraction. Remote Sensing, 14(19), 4744. https://doi.org/10.3390/rs14194744