Abstract
The availability of large amount of high-resolution aerial images, together with the recent advancement of deep convolutional neural networks (DCNNs) for extracting rich-and-hierarchical features from unstructured data, has propelled the automation progress of extracting roads from aerial images. Despite the superior performance of DCNNs, a common problem of choosing between the classification and segmentation DCNNs still remains. By comparing two state-of-the-art baseline classification/segmentation DCNNs in several industrial application scenarios, we illustrate that their relative performance may vary, leading to different choices. We also propose a strategy of fusing multiple pre-trained DCNNs and empirically discover that it guarantees superior results in all of the experimented scenarios, using far less development time. A few tools and pre-trained models (https://github.com/caolele/road-discovery) are open-sourced to facilitate research and engineering activities.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Inkscape-centerline-trace: https://github.com/fablabnbg/inkscape-centerline-trace.
- 2.
APLS API based on Dijkstra’s algorithm: https://github.com/CosmiQ/apls.
References
Aich S, van der Kamp W, Stavness I (2018) Semantic binary segmentation using convolutional networks without decoders. In: Proceedings of CVPR
Badrinarayanan V, Kendall A, Cipolla R (2017) Segnet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell 39(12):2481–2495
Chen LC, Papandreou G, Kokkinos I, Murphy K, Yuille AL (2018) Deeplab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Trans Pattern Anal Mach Intell 40(4):834–848
Chen LC, Yang Y, Wang J, Xu W, Yuille AL (2016) Attention to scale: scale-aware semantic image segmentation. In: Proceedings of CVPR, pp 3640–3649
Chen LC, Zhu Y, Papandreou G, Schroff F, Adam H (2018) Encoder-decoder with atrous separable convolution for semantic image segmentation. arXiv preprint arXiv:1802.02611
Christophe E, Inglada J (2007) Robust road extraction for high resolution satellite images. In: IEEE international conference on image processing, 2007, ICIP 2007, vol 5. IEEE, pp V–437
CosmiQWorks, DigitalGlobe, NVIDIA: SpaceNet on Amazon Web Services (AWS) Datasets: The SpaceNet Catalog. Accessed 3 June 2018. https://spacenetchallenge.github.io/datasets/datasetHomePage.html
Davydow A, Nikolenko S (2018) Land cover classification with superpixels and jaccard index post-optimization. In: The IEEE conference on computer vision and pattern recognition (CVPR) workshops, June 2018
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: The IEEE conference on computer vision and pattern recognition (CVPR) workshops, June 2018
Devlin J, Chang MW, Lee K, Toutanova K (2018) BERT: pre-training of deep bidirectional transformers for language understanding. arXiv:1810.04805
DSTL (2017) Kaggle satellite imagery feature detection. Accessed 3 June 2018. www.kaggle.com/c/dstl-satellite-imagery-feature-detection/data
Ghosh A, Ehrlich M, Shah S, Davis LS, Chellappa R (2018) Stacked U-Nets for ground material segmentation in remote sensing imagery. In: The IEEE conference on computer vision and pattern recognition (CVPR) workshops, June 2018
Han S, Mao H, Dally WJ (2016) Deep compression: compressing DNN with pruning, trained quantization and Huffman coding. In: Proceedings of ICLR
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of CVPR, pp 770–778
Hu J, Razdan A, Femiani JC, Cui M, Wonka P (2007) Road network extraction and intersection detection from aerial images by tracking road footprints. IEEE Trans Geosci Remote Sens 45(12):4144–4157
Huang X, Zhang L (2009) Road centreline extraction from high-resolution imagery based on multiscale structural features and support vector machines. Int J Remote Sens 30(8):1977–1987
Kuo TS, Tseng KS, Yan JW, Liu YC, Frank Wang YC (2018) Deep aggregation net for land cover classification. In: The IEEE conference on computer vision and pattern recognition (CVPR) workshops, June 2018
Lin G, Milan A, Shen C, Reid I (2017) RefineNet: multi-path refinement networks for high-resolution semantic segmentation. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 5168–5177
Mena JB, Malpica JA (2005) An automatic method for road extraction in rural and semi-urban areas starting from high resolution satellite imagery. Pattern Recogn Lett 26(9):1201–1220
Mnih V (2013) Machine learning for aerial image labeling. Ph.D. thesis, University of Toronto, Canada
Mnih V, Hinton GE (2010) Learning to detect roads in high-resolution aerial images. In: European conference on computer vision. Springer, pp 210–223
Mokhtarzade M, Zoej MV (2007) Road detection from high-resolution satellite images using artificial neural networks. Int J Appl Earth Obs Geoinf 9(1):32–40
Papandreou G, Chen LC, Murphy KP, Yuille AL (2015) Weakly-and semi-supervised learning of a deep convolutional network for semantic image segmentation. In: Proceedings of the IEEE international conference on computer vision, pp 1742–1750
Shelhamer E, Long J, Darrell T (2017) Fully convolutional networks for semantic segmentation. IEEE Trans Pattern Anal Mach Intell 39(4):640–651
Shen J, Lin X, Shi Y, Wong C (2008) Knowledge-based road extraction from high resolution remotely sensed imagery. In: Congress on image and signal processing, 2008, CISP’08, vol 4. IEEE, pp 608–612
Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. In: Proceedings of ICLR
Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A et al (2015) Going deeper with convolutions. In: Proceedings of CVPR. IEEE, pp 1–9
Wang J, Qin Q, Gao Z, Zhao J, Ye X (2016) A new approach to urban road extraction using high-resolution aerial image. ISPRS Int J Geo-Inf 5(7):114–126
Zhang Q, Couloigner I (2006) Benefit of the angular texture signature for the separation of parking lots and roads on high resolution multi-spectral imagery. Pattern Recogn Lett 27(9):937–946
Zhao H, Shi J, Qi X, Wang X, Jia J (2017) Pyramid scene parsing network. In: Proceedings of CVPR, pp 2881–2890
Zheng S, Jayasumana S, Romera-Paredes B, Vineet V, Su Z, Du D, Huang C, Torr PH (2015) Conditional random fields as recurrent neural networks. In: Proceedings of CVPR, pp 1529–1537
Zhou L, Zhang C, Wu M (2018) D-linknet: linknet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction. In: The IEEE conference on computer vision and pattern recognition (CVPR) workshops, June 2018
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Cao, L., Pan, X. (2020). Fusing Classification and Segmentation DCNNs for Road Feature Mining on Aerial Images. In: Popovich, V., Thill, JC., Schrenk, M., Claramunt, C. (eds) Information Fusion and Intelligent Geographic Information Systems . Advances in Geographic Information Science. Springer, Cham. https://doi.org/10.1007/978-3-030-31608-2_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-31608-2_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-31607-5
Online ISBN: 978-3-030-31608-2
eBook Packages: Earth and Environmental ScienceEarth and Environmental Science (R0)