Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery
Yıl: 2020 Cilt: 7 Sayı: 3 Sayfa Aralığı: 221 - 227 Metin Dili: İngilizce DOI: 10.30897/ijegeo.684951 İndeks Tarihi: 29-06-2021
Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery
Öz: Building extraction from high-resolution aerial imagery plays an important role in geospatial applications such as urban planning,telecommunication, disaster monitoring, navigation, updating geographic databases, and urban dynamic monitoring. Automaticbuilding extraction is a challenging task, as the buildings in different regions have different spectral and geometric properties.Therefore, the classical image processing techniques are not sufficient for automatic building extraction from high-resolution aerialimagery applications. Deep learning and semantic segmentation models, which have gained popularity in recent years, have beenused for automatic object extraction from high-resolution images. U-Net model, which was originally developed for biomedicalimage processing, was used for building extraction. The encoder part of the U-Net model has been modified with Vgg16,InceptionResNetV2, and DenseNet121 convolutional neural networks. Therefore, building extraction was performed using Vgg16 UNet, InceptionResNetV2 U-Net, and DenseNet121 U-Net models. In the fourth method, the results obtained from each U-Net modelwere combined in order to obtain the final result by maximum voting. This study aims to compare the performance of these fourmethods in building extraction from high-resolution aerial imagery. Images of Chicago from the Inria Aerial Image Labeling Datasetwere used in the study. The images used have 0.3 m spatial resolution, 8-bit radiometric resolution, and 3-band (red, green, and bluebands). Images consist of 36 tiles and they were divided into image subsets of 512x512 pixels. Thus, a total of 2715 image subsetswere formed. 80% of the image subsets (2172 image subset) were used as training and 20% (543 image subset) as testing. Toevaluate the accuracy of methods, the F1 score of the building class was employed. The F1 scores for building class have beencalculated as 0.866, 0.860, 0.856, and 0.877 on test images for U-Net Vgg16, U-Net InceptionResNetV2, U-Net DenseNet121, andmajority voting method, respectively.
Anahtar Kelime: Belge Türü: Makale Makale Türü: Araştırma Makalesi Erişim Türü: Erişime Açık
- Algancı, U., Sertel, E., Kaya, Ş. (2018). Determination of the Olive Trees with Object Based Classification of Pleiades Satellite Image. International Journal of Environment and Geoinformatics, 5(2), 132-139. DOI: 10.30897/ijegeo.396713.
- Bittner, K., Adam, F., Cui, S., Körner, M., and Reinartz, P. (2018). Building footprint extraction from VHR remote sensing images combined with normalized DSMs using fused fully convolutional networks. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(8), 2615-2629.
- Boonpook, W., Tan, Y., Ye, Y., Torteeka, P., Torsri, K. and Dong, S. (2018). A Deep Learning Approach on Building Detection from Unmanned Aerial VehicleBased Images in Riverbank Monitoring. Sensors, 18(11), 3921.
- Çelik, O., Gazioğlu, C. (2020). Coastline Difference Measurement (CDM) Method. International Journal of Environment and Geoinformatics, 7(1), 1-5. DOI: 10.30897/ijegeo.706792.
- Chen, X., Xiang, S., Liu, C. L., Pan, C. H. (2014). Vehicle detection in satellite images by hybrid deep convolutional neural networks. IEEE Geosci. Remote Sens. Lett., 11(10), 1797–1801.
- Cheng, G., Han, J. (2016). A survey on object detection in optical remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing, 117, 11-28.
- Chollet, F. (2020). Keras. Retrieved 27 April 2020 from https://github.com/fchollet/keras/
- Dervisoglu, A, Bilgilioğlu, B., Yağmur, N. (2020). Comparison of Pixel-Based and Object-Based Classification Methods in Determination of Wetland Coastline. International Journal of Environment and Geoinformatics, 7(2), 213-220. DOI: 10.30897/ijegeo.713307.
- Esetlili, M., Bektas Balcik, F., Balik Sanli, F., Kalkan, K., Ustuner, M., Goksel, Ç., Gazioğlu, C., Kurucu, Y. (2018). Comparison of Object and Pixel-Based Classifications for Mapping Crops Using Rapideye Imagery: A Case Study of Menemen Plain, Turkey. International Journal of Environment and Geoinformatics, 5(2), 231-243. DOI: 10.30897/ijegeo.442002.
- Ghanea, M., Moallem, P., Momeni, M. (2016). Building extraction from high-resolution satellite images in urban areas: recent methods and strategies against significant challenges. International journal of remote sensing, 37(21), 5234-5248.
- Hu, W., Huang, Y., Wei, L., Zhang, F., Li, H. (2015). Deep convolutional neural networks for hyperspectral image classification, Journal of Sensors, 2015, Article ID 258619, 1–12.
- Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K. Q. (2017). Densely Connected Convolutional Networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, 4700-4708.
- Huang, W., Xiao, L., Wei, Z., Liu, H., Tang, S. (2015). A new pan sharpening method with deep neural networks. IEEE Geosci. Remote Sens. Lett., 12(5), 1037–1041.
- Hui, J., Du, M., Ye, X., Qin, Q., and Sui, J. (2018). Effective building extraction from high-resolution remote sensing images with multitask driven deep neural network. IEEE Geoscience and Remote Sensing Letters, 16(5), 786-790.
- Ji, S., Wei, S., and 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.
- Ji, S., Wei, S., and Lu, M. (2019). A scale robust convolutional neural network for automatic building extraction from aerial and satellite imagery. International journal of remote sensing, 40(9), 3308-3322.
- Li, L., Liang, J., Weng, M., and Zhu, H. (2018). A multiple-feature reuse network to extract buildings from remote sensing imagery. Remote Sensing, 10(9),1350.
- Li, X., Yao, X., Fang, Y. (2018). Building-A-Nets: Robust Building Extraction From High-Resolution Remote Sensing Images With Adversarial Networks. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 99, 1-8.
- Lin, J., Jing, W., Song, H., Chen, G. (2019). ESFNet: Efficient Network for Building Extraction From High-Resolution Aerial Images. IEEE Access, 7, 54285-54294.
- Liu, H., Luo, J., Huang, B., Hu, X., Sun, Y., Yang, Y., ... and Zhou, N. (2019). DE-Net: Deep Encoding Network for Building Extraction from HighResolution Remote Sensing Imagery. Remote Sensing, 11(20), 2380.
- Liu, P., Liu, X., Liu, M., Shi, Q., Yang, J., Xu, X., and Zhang, Y. (2019). Building footprint extraction from high-resolution images via spatial residual inception convolutional neural network. Remote Sensing, 11(7), 830.
- Liu, Y., Gross, L., Li, Z., Li, X., Fan, X., and Qi, W. (2019). Automatic building extraction on highresolution remote sensing imagery using deep convolutional encoder-decoder with spatial pyramid pooling. IEEE Access, 7, 128774-128786.
- Long, J., Shelhamer, E., Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), 3431-3440.
- Lu, T., Ming, D., Lin, X., Hong, Z., Bai, X., and Fang, J. (2018). Detecting building edges from high spatial resolution remote sensing imagery using richer convolution features network. Remote Sensing, 10(9), 1496.
- Maggiori, E., Tarabalka, Y., Charpiat, G., Alliez, P. (2017). Can semantic labeling methods generalize to any city? the inria aerial image labeling benchmark. In 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 3226-3229.
- Pan, X., Yang, F., Gao, L., Chen, Z., Zhang, B., Fan, H., and Ren, J. (2019). Building extraction from highresolution aerial imagery using a generative adversarial network with spatial and channel attention mechanisms. Remote Sensing, 11(8), 917.
- Ronneberger, O., Fischer, P., Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, 234-241.
- Schuegraf, P., and Bittner, K. (2019). Automatic Building Footprint Extraction from Multi-Resolution Remote Sensing Images Using a Hybrid FCN. ISPRS International Journal of Geo-Information, 8(4), 191.
- Simonyan, K., Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition, arXiv preprint, arXiv:1409.1556.
- Sun, G., Huang, H., Zhang, A., Li, F., Zhao, H., and Fu, H. (2019). Fusion of multiscale convolutional neural networks for building extraction in very highresolution images. Remote Sensing, 11(3), 227.
- Szegedy, C., Ioffe, S., Vanhoucke, V., Alemi, A. A. (2017). Inception-v4, inception-resnet and the impact of residual connections on learning. In Thirty-First AAAI Conference on Artificial Intelligence. 4278-4284.
- Xu, Y., Wu, L., Xie, Z., and Chen, Z. (2018). Building extraction in very high resolution remote sensing imagery using deep learning and guided filters. Remote Sensing, 10(1), 144.
- Yang, H., Wu, P., Yao, X., Wu, Y., Wang, B., Xu, Y. (2018). Building extraction in very high resolution imagery by dense-attention networks. Remote Sensing, 10(11), 1768.
- Yi, Y., Zhang, Z., Zhang, W., Zhang, C., Li, W., and Zhao, T. (2019). Semantic Segmentation of urban buildings from vhr remote sensing imagery using a deep convolutional neural network. Remote Sensing, 11(15), 1774.
- Zeiler, M. D. (2012). Adadelta: an adaptive learning rate method. arXiv preprint, arXiv:1212.5701.
- Zhang, F., Du, B., Zhang, L. (2016). Scene classification via a gradient boosting random convolutional network framework. IEEE Trans. Geosci. Remote Sens., 54(3), 1793–1802.
- Zhang, L., Wu, J., Fan, Y., Gao, H., and Shao, Y. (2020). An Efficient Building Extraction Method from High Spatial Resolution Remote Sensing Images Based on Improved Mask R-CNN. Sensors, 20(5), 1465.
| APA | Erdem F, Avdan U (2020). Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. , 221 - 227. 10.30897/ijegeo.684951 |
| Chicago | Erdem Fırat,Avdan Ugur Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. (2020): 221 - 227. 10.30897/ijegeo.684951 |
| MLA | Erdem Fırat,Avdan Ugur Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. , 2020, ss.221 - 227. 10.30897/ijegeo.684951 |
| AMA | Erdem F,Avdan U Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. . 2020; 221 - 227. 10.30897/ijegeo.684951 |
| Vancouver | Erdem F,Avdan U Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. . 2020; 221 - 227. 10.30897/ijegeo.684951 |
| IEEE | Erdem F,Avdan U "Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery." , ss.221 - 227, 2020. 10.30897/ijegeo.684951 |
| ISNAD | Erdem, Fırat - Avdan, Ugur. "Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery". (2020), 221-227. https://doi.org/10.30897/ijegeo.684951 |
| APA | 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. 10.30897/ijegeo.684951 |
| Chicago | Erdem Fırat,Avdan Ugur Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. International Journal of Environment and Geoinformatics 7, no.3 (2020): 221 - 227. 10.30897/ijegeo.684951 |
| MLA | Erdem Fırat,Avdan Ugur Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. International Journal of Environment and Geoinformatics, vol.7, no.3, 2020, ss.221 - 227. 10.30897/ijegeo.684951 |
| AMA | Erdem F,Avdan U Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. International Journal of Environment and Geoinformatics. 2020; 7(3): 221 - 227. 10.30897/ijegeo.684951 |
| Vancouver | Erdem F,Avdan U Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery. International Journal of Environment and Geoinformatics. 2020; 7(3): 221 - 227. 10.30897/ijegeo.684951 |
| IEEE | Erdem F,Avdan U "Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery." International Journal of Environment and Geoinformatics, 7, ss.221 - 227, 2020. 10.30897/ijegeo.684951 |
| ISNAD | Erdem, Fırat - Avdan, Ugur. "Comparison of Different U-Net Models for Building Extraction from High-Resolution Aerial Imagery". International Journal of Environment and Geoinformatics 7/3 (2020), 221-227. https://doi.org/10.30897/ijegeo.684951 |
