Abstract
This study presents an innovative approach for automatic road detection with deep learning, employing fusion strategies to utilize both lower-resolution satellite imagery and GPS trajectory data, a concept never explored before. We rigorously investigate both early and late fusion strategies and assess deep learning-based road detection performance using different fusion settings. Our extensive ablation studies evaluate the efficacy of our framework under diverse model architectures, loss functions, and geographic domains (Istanbul and Montreal). For an unbiased and complete evaluation of road detection results, we use both region-based and boundary-based evaluation metrics for road segmentation. The outcomes reveal that the ResUnet model outperforms U-Net and D-Linknet in road extraction tasks, achieving superior results over the benchmark study using low-resolution Sentinel-2 data. This research not only contributes to the field of automatic road detection but also offers novel insights into the utilization of data fusion methods in diverse applications.
Similar content being viewed by others
Availability of data and material
The code of the experiments and the data used in the experiments of this study are made available online, and related website information is shared within the article.
Notes
The pre-processed data can be downloaded from following URL: https://github.com/nagellette/sentinel_traj_nn/blob/master/Data.md
The implementations of the methods and experiments can be downloaded from the following URL: https://github.com/nagellette/sentinel_traj_nn
References
Jiao C, Heitzler M, Hurni L (2021) A survey of road feature extraction methods from raster maps. Trans GIS 25(6):2734–2763
Zhou L, Zhang C, Wu M (2018) D-linknet: Linknet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction, vol. 2018-June, pp 192–196. IEEE Computer Society, Salt Lake City https://doi.org/10.1109/CVPRW.2018.00034
Demir I, Koperski K, Lindenbaum D, Pang G, Huang J, Basu S, Hughes F, Tuia D, Raskar R (2018) Deepglobe 2018: A challenge to parse the earth through satellite images. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp 172–17209. https://doi.org/10.1109/CVPRW.2018.00031
Liu P, Wang Q, Yang G, Li L, Zhang H (2022) Survey of road extraction methods in remote sensing images based on deep learning. PFG - J Photogramm, Remote Sens Geoinf Sci 90(2):135–159. https://doi.org/10.1007/s41064-022-00194-z
Mnih V, Hinton GE (2010) Learning to detect roads in high-resolution aerial images. In: Daniilidis K, Maragos P, Paragios N (eds) Computer Vision - ECCV 2010. Springer, Berlin, Heidelberg, pp 210–223
Zhang Z, Liu Q, Wang Y (2018) Road extraction by deep residual U-Net. IEEE Geosci Remote Sens Lett 15(5):749–753. https://doi.org/10.1109/LGRS.2018.2802944
Sun T, Chen Z, Yang W, Wang Y (2018) Stacked U-Nets with multi-output for road extraction. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). pp 187–1874. https://doi.org/10.1109/CVPRW.2018.00033
Ayala C, Aranda C, Galar M (2021) Towards fine-grained road maps extraction using Sentinel-2 imagery. ISPRS Ann Photogramm, Remote Sens Spatial Inf Sci V-3-2021:9–14. https://doi.org/10.5194/isprs-annals-V-3-2021-9-2021
Johnson N, Treible W, Crispell D (2022) OpenSentinelMap: A large-scale land use dataset using OpenStreetMap and Sentinel-2 imagery. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). pp 1332–1340. https://doi.org/10.1109/CVPRW56347.2022.00139
Edelkamp S, Schrödl S (2003). In: Klein R, Six H-W, Wegner L (eds) Route Planning and Map Inference with Global Positioning Traces. Springer, Berlin, Heidelberg, pp 128–151
Biagioni J, Eriksson J (2012) Map inference in the face of noise and disparity. In: Proceedings of the 20th international conference on advances in geographic information systems. SIGSPATIAL ’12. Association for Computing Machinery, New York, NY, USA, pp 79–88. https://doi.org/10.1145/2424321.2424333. https://doi.org/10.1145/2424321.2424333
Karagiorgou S, Pfoser D (2012) On vehicle tracking data-based road network generation. In: Proceedings of the 20th international conference on advances in geographic information systems. SIGSPATIAL ’12. Association for Computing Machinery, New York, NY, USA, pp 89–98. https://doi.org/10.1145/2424321.2424334. https://doi.org/10.1145/2424321.2424334
Tang J, Deng M, Huang J, Liu H, Chen X (2019) An automatic method for detection and update of additive changes in road network with gps trajectory data. ISPRS Int J Geo-Inf 8(9). https://doi.org/10.3390/ijgi8090411
Sun T, Di Z, Che P, Liu C, Wang Y (2019) Leveraging crowdsourced gps data for road extraction from aerial imagery. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Liu L, Yang Z, Li G, Wang K, Chen T, Lin L (2022) Aerial images meet crowdsourced trajectories: A new approach to robust road extraction. IEEE Trans Neural Netw Learn Syst 1–15. https://doi.org/10.1109/TNNLS.2022.3141821
Wu H, Zhang H, Zhang X, Sun W, Zheng B, Jiang Y (2020) DeepDualMapper: A gated fusion network for automatic map extraction using aerial images and trajectories. Proc AAAI Conf Artif Intell 34:1037–1045
Yang J, Ye X, Wu B, Gu Y, Wang Z, Xia D, Huang J (2022) DuARE: Automatic road extraction with aerial images and trajectory data at baidu maps. In: Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining. KDD ’22. Association for Computing Machinery, New York, NY, USA , pp 4321–4331. https://doi.org/10.1145/3534678.3539029 . https://doi.org/10.1145/3534678.3539029
Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional networks for biomedical image segmentation. In: Navab N, Hornegger J, Wells WM, Frangi AF (eds) Medical image computing and computer-assisted intervention - MICCAI 2015. Springer, Cham, pp 234–241
Wu Z, Zhang J, Zhang L, Liu X, Qiao H (2022) Bi-HRNet: A road extraction framework from satellite imagery based on node heatmap and bidirectional connectivity. Remote Sens 14(7). https://doi.org/10.3390/rs14071732
Xie Y, Miao F, Zhou K, Peng J (2019) HsgNet: A road extraction network based on global perception of high-order spatial information. ISPRS Int J Geo-Inf 8(12). https://doi.org/10.3390/ijgi8120571
Dai L, Zhang G, Zhang R (2023) RADANet: Road augmented deformable attention network for road extraction from complex high-resolution remote-sensing images. IEEE Trans Geosci Remote Sens 61:1–13. https://doi.org/10.1109/TGRS.2023.3237561
Yang M, Yuan Y, Liu G (2022) SDUNet: Road extraction via spatial enhanced and densely connected unet. Pattern Recognit 126:108549. https://doi.org/10.1016/j.patcog.2022.108549
Li S, Liao C, Ding Y, Hu H, Jia Y, Chen M, Xu B, Ge X, Liu T, Wu D (2022) Cascaded residual attention enhanced road extraction from remote sensing images. ISPRS Int J Geo-Inf 11(1). https://doi.org/10.3390/ijgi11010009
Luo L, Wang J-X, Chen S-B, Tang J, Luo B (2022) BDTNet: Road extraction by bi-direction transformer from remote sensing images. IEEE Geosci Remote Sens Lett 19:1–5. https://doi.org/10.1109/LGRS.2022.3183828
Jiang X, Li Y, Jiang T, Xie J, Wu Y, Cai Q, Jiang J, Xu J, Zhang H (2022) RoadFormer: Pyramidal deformable vision transformers for road network extraction with remote sensing images. Int J Appl Earth Obs Geoinf 113:102987. https://doi.org/10.1016/j.jag.2022.102987
Tao J, Chen Z, Sun Z, Guo H, Leng B, Yu Z, Wang Y, He Z, Lei X, Yang J (2023) Seg-Road: A segmentation network for road extraction based on transformer and cnn with connectivity structures. Remote Sens 15(6). https://doi.org/10.3390/rs15061602
Ayala C, Sesma R, Aranda C, Galar M (2021) A deep learning approach to an enhanced building footprint and road detection in high-resolution satellite imagery. Remote Sens 13(16). https://doi.org/10.3390/rs13163135
Li P, He X, Qiao M, Miao D, Cheng X, Song D, Chen M, Li J, Zhou T, Guo X, Yan X, Tian Z (2021) Exploring multiple crowdsourced data to learn deep convolutional neural networks for road extraction. Int J Appl Earth Obs Geoinf 104:102544. https://doi.org/10.1016/j.jag.2021.102544
Zhou K, Xie Y, Gao Z, Miao F, Zhang L (2021) Funet: A novel road extraction network with fusion of location data and remote sensing imagery. ISPRS Int J Geo-Inf 10(1). https://doi.org/10.3390/ijgi10010039
Eftelioglu E, Garg R, Kango V, Gohil C, Chowdhury AR (2022) RING-Net: Road inference from gps trajectories using a deep segmentation network. In: Proceedings of the 10th ACM SIGSPATIAL international workshop on analytics for big geospatial data. BigSpatial ’22. Association for Computing Machinery, New York, NY, USA, pp 17–26. https://doi.org/10.1145/3557917.3567617 . https://doi.org/10.1145/3557917.3567617
Gao L, Wang J, Wang Q, Shi W, Zheng J, Gan H, Lv Z, Qiao H (2021) Road extraction using a dual attention dilated-linknet based on satellite images and floating vehicle trajectory data. IEEE J Sel Top Appl Earth Obs Remote Sens 14:10428–10438. https://doi.org/10.1109/JSTARS.2021.3116281
Zhang Y, Sidibé D, Morel O, Mériaudeau F (2021) Deep multimodal fusion for semantic image segmentation: A survey. Image Vision Comput 105:104042. https://doi.org/10.1016/j.imavis.2020.104042
Goodfellow I, Bengio Y, Courville A (2016) Deep Learning. MIT Press, Cambridge . http://www.deeplearningbook.org
Maxwell AE, Warner TA, Guillén LA (2021a) Accuracy assessment in convolutional neural network-based deep learning remote sensing studies–part 1: Literature review. Remote Sens 13(13). https://doi.org/10.3390/rs13132450
Maxwell AE, Warner TA, Guillén LA (2021b) Accuracy assessment in convolutional neural network-based deep learning remote sensing studies–part 2: Recommendations and best practices. Remote Sens 13(13). https://doi.org/10.3390/rs13132591
Etten AV (2020) City-scale road extraction from satellite imagery v2: Road speeds and travel times. In: 2020 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, Snowmass Village. https://doi.org/10.1109/wacv45572.2020.9093593 . https://doi.org/10.1109/wacv45572.2020.9093593
Etten AV, Shermeyer J, Hogan D, Weir N, Lewis R (2020) Road network and travel time extraction from multiple look angles with spacenet data. In: IGARSS 2020 - 2020 IEEE international geoscience and remote sensing symposium. IEEE, Virtual Symposium. https://doi.org/10.1109/igarss39084.2020.9324091
Maier-Hein L, Reinke A, Godau P, Tizabi MD, Büttner F, Christodoulou E, Glocker B, Isensee F, Kleesiek J, Kozubek M, Reyes M, Riegler MA, Wiesenfarth M, Kavur E, Sudre CH, Baumgartner M, Eisenmann M, Heckmann-Nötzel D, Rädsch AT, Acion L, Antonelli M, Arbel T, Bakas S, Benis A, Blaschko M, Cardoso MJ, Cheplygina V, Cimini BA, Collins GS, Farahani K, Ferrer L, Galdran A, Ginneken B, Haase R, Hashimoto DA, Hoffman MM, Huisman M, Jannin P, Kahn CE, Kainmueller D, Kainz B, Karargyris A, Karthikesalingam A, Kenngott H, Kofler F, Kopp-Schneider A, Kreshuk A, Kurc T, Landman BA, Litjens G, Madani A, Maier-Hein K, Martel AL, Mattson P, Meijering E, Menze B, Moons KGM, Müller H, Nichyporuk B, Nickel F, Petersen J, Rajpoot N, Rieke N, Saez-Rodriguez J, Sánchez CI, Shetty S, Smeden M, Summers RM, Taha AA, Tiulpin A, Tsaftaris SA, Calster BV, Varoquaux G, Jäger PF (2023) Metrics reloaded: Pitfalls and recommendations for image analysis validation. arXiv:2206.01653
Ozturk O, Isik MS, Sariturk B, Seker DZ (2022) Generation of istanbul road data set using google map api for deep learning-based segmentation. Int J Remote Sens 43:2793–2812. https://doi.org/10.1080/01431161.2022.2068989
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Ciampiconi L, Elwood A, Leonardi M, Mohamed A, Rozza A (2023) A survey and taxonomy of loss functions in machine learning
Jadon S (2020) A survey of loss functions for semantic segmentation. In: 2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB). pp 1–7. https://doi.org/10.1109/CIBCB48159.2020.9277638
Lin T-Y, Goyal P, Girshick R, He K, Dollár P (2020) Focal loss for dense object detection. IEEE Trans Pattern Anal Mach Intell 42(2):318–327. https://doi.org/10.1109/TPAMI.2018.2858826
Cheng B, Girshick R, Dollar P, Berg AC, Kirillov A (2021) Boundary iou: Improving object-centric image segmentation evaluation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp 15334–15342
Istanbul Buyuksehir Belediyesi (2020) IBB ISTAÇ Araçlarinin Anlik Konum ve Hiz Bilgileri . https://data.ibb.gov.tr/dataset/ibb-istac-araclarinin-anlik-konum-ve-hiz-bilgileri. [Accessed 23-Jun-2020]
Ville de Montréal (2021) VMTL-MTL-Trajet. https://www.donneesquebec.ca/recherche/fr/dataset/vmtl-mtl-trajet. [Accessed 12-Apr-2023]
European Space Agency (2021) Sentinel-2 MSI - MultiSpectral Instrument User Guide. https://sentinels.copernicus.eu/web/sentinel/user-guides/sentinel-2-msi. [Accessed 12-Apr-2023]
OpenStreetMap Contributors (2021) OpenStreetMap. https://www.openstreetmap.org/. [Accessed 12-Apr-2023]
Yuan M, Liu Z, Wang F, Jin F (2019) Rethinking labelling in road segmentation. Int J Remote Sens 40(22):8359–8378. https://doi.org/10.1080/01431161.2019.1608393
Gengec NE, Tari E (2021) Performance evaluation of gps trajectory rasterization methods. In: Computational Science and Its Applications – ICCSA 2021: 21st International Conference, Cagliari, Italy, September 13–16, 2021, Proceedings, Part I. Springer, Berlin, Heidelberg, pp 3–17. https://doi.org/10.1007/978-3-030-86653-2_1
Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, Corrado GS, Davis A, Dean J, Devin M, Ghemawat S, Goodfellow I, Harp A, Irving G, Isard M, Jia Y, Jozefowicz R, Kaiser L, Kudlur M, Levenberg J, Mané D, Monga R, Moore S, Murray D, Olah C, Schuster M, Shlens J, Steiner B, Sutskever I, Talwar K, Tucker P, Vanhoucke V, Vasudevan V, Viégas F, Vinyals O, Warden P, Wattenberg M, Wicke M, Yu Y, Zheng X (2015) TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. Software available from tensorflow.org. https://www.tensorflow.org/
TensorFlow Addons Contributors (2021) TensorFlow Addons. https://github.com/tensorflow/addons
Perkiö J, Hyvärinen A (2009) Modelling image complexity by independent component analysis, with application to content-based image retrieval. In: Alippi C, Polycarpou M, Panayiotou C, Ellinas G (eds) Artificial Neural Networks - ICANN 2009. Springer, Berlin, Heidelberg, pp 704–714
Haralick RM, Shanmugam K, Dinstein I (1973) Textural features for image classification. IEEE Trans Syst, Man, Cybernet SMC-3(6):610–621. https://doi.org/10.1109/TSMC.1973.4309314
Rahane AA, Subramanian A (2020) Measures of complexity for large scale image datasets. In: 2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). pp 282–287. https://doi.org/10.1109/ICAIIC48513.2020.9065274
Bilgi S, Gulnerman AG, Arslanoglu B, Karaman H, Ozturk O (2019) Complexity measures of sports facilities allocation in urban area by metric entropy and public demand compatibility. Int J Eng Geosci 4(3):141–148. https://doi.org/10.26833/ijeg.540180
scikit-image (2021) Shannon entropy. https://scikit-image.org/docs/stable/api/skimage.measure.html#skimage.measure.shannon_entropy. [Online; accessed 12-Apr-2023]
scikit-image (2021b) Gray-level co-occurrence matrix properties. https://scikit-image.org/docs/stable/api/skimage.feature.html#skimage.feature.graycoprops. [Online; accessed 12-Apr-2023]
Mapillary (2023) Image API documentation. https://www.mapillary.com/developer/api-documentation#image. [Accessed 16-Nov-2023]
Open Street Map (2023) Public GPS Traces. https://www.openstreetmap.org/traces/. [Accessed 16-Nov-2023]
Acknowledgements
The numerical calculations reported in this paper were fully performed at TUBITAK ULAKBIM, High Performance and Grid Computing Center (TRUBA resources). The authors would like to thank the Istanbul Metropolitan Municipality and City of Montreal for the GPS trajectory dataset, the European Space Agency (ESA) for Sentinel-2 data, and the OpenStreetMap Foundation and OpenStreetMap Contributors for OpenStreetMap data. This study is part of the Ph.D. thesis conducted at Istanbul Technical University by the first author. The authors would like to thank the Ph.D. thesis advancement monitoring committee members, Gulsen Kaya Taskin and Taskin Kavzoglu.
Funding
No funding was received to assist with the preparation of this manuscript.
Author information
Authors and Affiliations
Contributions
Necip Enes Gengec contributed to the manuscript by conceptualizing the topic, running experiments, writing the main manuscript, and preparing figures and tables. Ergin Tari and Ulas Bagci provided valuable guidance and expertise in the development of the methodology and execution of the experiments. All authors actively participated in reviewing the manuscript and contributed to its improvement through discussions and suggestions.
Corresponding author
Ethics declarations
Conflicts of interest/Competing interests
The authors declare no conflict of interest or competing interest.
Additional information
Communicated by: H. Babaie.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Gengec, N.E., Tari, E. & Bagci, U. AI powered road network prediction with fused low-resolution satellite imagery and GPS trajectory. Earth Sci Inform 17, 1013–1029 (2024). https://doi.org/10.1007/s12145-023-01201-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12145-023-01201-6