Abstract
Deep learning techniques have recently been experimented for map generalization. Although promising, these experiments raise new problems regarding the evaluation of the output images. Traditional map generalization evaluation cannot directly be applied to the results in a raster format. Additionally, the internal evaluation used by deep learning models is mostly based on the realism of images and the accuracy of pixels, and none of these criteria is sufficient to evaluate a generalization process. Finally, deep learning processes tend to hide the causal mechanisms and do not always guarantee a result that follows cartographic principles. In this article, we propose a method to adapt constraint-based evaluation to the images generated by deep learning models. We focus on the use case of mountain road generalization, and detail seven raster-based constraints, namely, clutter, coalescence reduction, smoothness, position preservation, road connectivity preservation, noise absence, and color realism constraints. These constraints can contribute to current studies on deep learning-based map generalization, as they can help guide the learning process, compare different models, validate these models, and identify remaining problems in the output images. They can also be used to assess the quality of training examples.
Similar content being viewed by others
Data and Code availability
Data and code presented in the submitted work are available in the DeepMapGen repository, https://github.com/umrlastig/DeepMapGen.
References
Bard, S (2004) Quality assessment of cartographic generalisation. Transactions in GIS 8(1):63–81 ISSN: 1467-9671. https://doi.org/10.1111/j.1467-9671.2004.00168.x
Beard KM (1991) Constraints on rule formation. In: Buttenfield Barbara, McMaster Robert (eds) Map generalization. Longman Pages, pp 121–135
Bravo MJ, Farid H, (2008) A scale invariant measure of clutter. J Vision 8(1):23. https://jov.arvojournals.org/article.aspx?articleid=2122173. Publisher: The Association for Research in Vision and Ophthalmology
Burghardt D, Schmid S, Stoter JE (2007) Investigations on cartographic constraint formalisation. In: ICC 2007 : proceedings of the workshop of the ICA commission on generalization and multiple representation, August 2-3 at the 23nd international cartographic conference ICC: Cartography for everyone and for you, 4-10 August 2007. ICC, Moscow, Russia, p 16
Courtial A, El Ayedi A, Touya G, Zhang X (2020a) Exploring the potential of deep learning segmentation for mountain roads generalisation. ISPRS Int J Geo-Inf 9(5):338. https://doi.org/10.3390/ijgi9050338. Number: 5 Publisher: Multidisciplinary Digital Publishing Institute
Courtial A, Touya G, Zhang X (2020b) Constraint based evaluation of generalized images generated by deep learning. In: 23rd ICA workshop on map generalisation and multiple representation. Delft, Netherlands, ICA. https://hal.archives-ouvertes.fr/hal-02995412
Courtial A, Touya G, Zhang X (2021a) Generative adversarial networks to generalise urban area in topographic maps. In: The international archives of the photogrammetry, remote sensing and spatial information sciences, vol XLIII-B4-2021. Copernicus GmbH, pp 15–22. ISSN: 1682-1750. https://doi.org/10.5194/isprs-archives-XLIII-B4-2021-15-2021
Courtial A, Touya G, Zhang X (2021b) Can graph convolution networks learn spatial relations? Abstracts of the ICA 3:1–2. https://doi.org/10.5194/ica-abs-3-60-2021
Courtial A, Touya G, Zhang X (2022) AlpineBends – A Benchmark for Deep Learning-Based Generalisation. Abstracts of the ICA 4:1–2. https://doi.org/10.5194/ica-abs-4-1-2022
Du J, Wu F, Xing R, Gong X, Yu L (2021) Segmentation and sampling method for complex polyline generalization based on a generative adversarial network. Geocarto Int 0(0):1–23. ISSN 1010-6049. https://doi.org/10.1080/10106049.2021.1878288. Publisher: Taylor & Francis
Duchêne C (2014) Individual road generalisation in the 1997–2000 AGENT European project. Technical report, IGN, COGIT lab, Saint-Mandé, France
Dumont M, Touya G, Duchêne C (2016) Assessing the variation of visual complexity in multi-scale maps with clutter measures. ICA Workshop on Generalisation and Multiple Representation
Feng Y, Thiemann F, Sester M (2019) Learning cartographic building generalization with deep convolutional neural networks. International Journal of Geo-Information
Fu H, Gong M, Wang C, Batmanghelich K, Zhang K, Tao D (2019) Geometry-consistent generative adversarial networks for one-sided unsupervised domain mapping. In: 2019 IEEE/CVF conference on computer vision and pattern recognition (CVPR). IEEE, Long Beach, CA, USA. pp 2422–2431. ISBN 978-1-72813-293-8 https://doi.org/10.1109/CVPR.2019.00253
Galanda M (2003) Modelling constraints for polygon generalisation. In: Proceedings of 5th workshop on progress in automated map generalisation. ICA, event-place: Paris, France
Isola P, Zhu JY, Zhou T, Efros AA (2016) Image-to-image translation with conditional adversarial networks. arXiv:1611.07004
Kang Y, Gao S, Roth R (2019) Transferring multiscale map style using generative adversarial network
Mackaness WA, Edwards G (2002) The importance of modelling pattern and structure in automated map generalisation. In: Proceedings of the Joint ISPRS/ICA workshop on multi-scale representations of spatial data. pp 7–8
Mackaness WA, Ruas A (2007) Chapter \(5\) - evaluation in the map generalisation process. In: Mackaness WA, Ruas A, Sarjakoski LT (eds) Generalisation of geographic information, International Cartographic Association. Elsevier Science B.V, Amsterdam, pp 89–111. ISBN 978-0-08-045374-3. https://doi.org/10.1016/B978-008045374-3/50007-7
Mustière S (1998) GALBE: Adaptive generalisation - the need for an adaptive process for automated generalisation, an example on roads. In: 1st GIS’PlaNet conference, Lisbon, Portugal
Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional networks for biomedical image segmentation. pp 234–241. arXiv:1505.04597
Rosenholtz R, Li Y, Nakano L (2007) Measuring visual clutter. Journal of Vision 7:17.1-22. https://doi.org/10.1167/7.2.17
Ruas, A (1999) Modèle de généralisation de données géographiques à base de contraintes et d’autonomie. phdthesis, Université de Marne la Vallée
Sharma G, Wu W, Dalal EN (2005) The CIEDE2000 color-difference formula: Implementation notes, supplementary test data, and mathematical observations. Color Res Appl 30(1):21–30 ISSN 0361-2317, 1520-6378. https://doi.org/10.1002/col.20070
Skopeliti A, Tsoulos L (2001) A methodology for the assessment of generalization quality. In: Fourth workshop on progress in automated map generalization. pp 2–4
Stoter J, Burghardt D, Duchêe C, Baella B, Bakker N, Blok C, Pla M, Regnauld N, Touya G, Schmid S (2009) Methodology for evaluating automated map generalization in commercial software. Comput Environ Urban Syst 33(5): 311–324. ISSN 0198-9715. https://doi.org/10.1016/j.compenvurbsys.2009.06.002
Stoter J, Zhang X, Stigmar H, Harrie L (2014) Evaluation in generalisation. In: Burghardt D, Duchêne C, Mackaness W (eds) Abstracting geographic information in a data rich World, Lecture Notes in Geoinformation and Cartography.Springer International Publishing, pp 259–297. https://doi.org/10.1007/978-3-319-00203-3_9
Touya G, Duchêne C, Ruas A (2010) Collaborative generalisation: formalisation of generalisation knowledge to orchestrate different cartographic generalisation processes. In: Fabrikant SI, Reichenbacher T, van Kreveld M, Schlieder C (eds) Geographic information science. lecture notes in computer science. Springer, Berlin, Heidelberg, pp 264–278. ISBN 978-3-642-15300-6 https://doi.org/10.1007/978-3-642-15300-6_19
Touya G (2012) Social welfare to assess the global legibility of a generalized map. Geographic Information Science
Touya G, Decherf B, Lalanne M, Dumont M (2015) Comparing image-based methods for assessing visual clutter in generalized maps. ISPRS Annals of Photogrammetry, Remote Sensing andSpatialInformationSciences, II-3/W5. https://doi.org/10.5194/isprsannals-II-3-W5-227-2015
Touya G, Zhang X, Lokhat, (2019) Is deep learning the new agent for map generalization? Int J Cartogr 5(2–3):142–157. ISSN 2372–9333. https://doi.org/10.1080/23729333.2019.1613071. Publisher: Taylor & Francis
Touya G, Lokhat I (2020) Deep learning for enrichment of vector spatial databases: Application to highway interchange. ACM Trans Spatial Algorithms Syst 6(3).ISSN 2374-0353. https://doi.org/10.1145/3382080
Werder S (2009) Formalization of spatial constraints. In: 12th AGILE international conference on geographic information science. p 13
Yan X, Ai T, Yang M, Tong X (2020) Graph convolutional autoencoder model for the shape coding and cognition of buildings in maps. Int J Geogr Inf Sci 0 (0): 1–23. ISSN 1365-8816. Publisher: Taylor & Francis. https://doi.org/10.1080/13658816.2020.1768260
Zhang X (2012) Automated evaluation of generalized topographic maps. PhD thesis, University of Twente
Zhang X, Stoter J, Ai T, Kraak M-J, Molenaar Ma (2013) Automated evaluation of building alignments in generalized maps. Int J Geogr Inf Sci 27 (8): 1550–1571. ISSN 1365-8816. Publisher: Taylor & Francis. https://doi.org/10.1080/13658816.2012.758264
Zhu J-Y, Park T, Isola P, Efros AA (2017) Unpaired image-to-image translation using cycle-consistent adversarial networks. In: 2017 IEEE International Conference on Computer Vision (ICCV). IEEE, Venice, pp 2242–2251. ISBN 978-1-5386-1032-9. https://doi.org/10.1109/ICCV.2017.244
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Informed Consent
This research involved no human participants.
Ethical Approval
This research involved no human participants.
Conflict of Interest
The authors declare no competing interests.
Additional information
This article belongs to the Topical Collection: Automated map generalization: emerging techniques and new trends
Guest Editors: Xiang Zhang, Guillaume Touya, Martijn Meijers.
Rights and permissions
About this article
Cite this article
Courtial, A., Touya, G. & Zhang, X. Constraint-Based Evaluation of Map Images Generalized by Deep Learning. J geovis spat anal 6, 13 (2022). https://doi.org/10.1007/s41651-022-00104-2
Accepted:
Published:
DOI: https://doi.org/10.1007/s41651-022-00104-2