Abstract
This paper deals with automatic image segmentation in poorly resourced areas. We concentrate on map content segmentation in historical maps as an example of such a domain. In such cases, conventional computer vision (CV) approaches fail in unexpected unique regions such as map content area exceeding the map frame, while deep learning methods lack boundary localization accuracy. Therefore, we propose an efficient approach that combines conventional CV techniques with deep learning and practically eliminates their drawbacks. To do so, we redefine the learning objective of a simple fully convolutional network to make the training easier and the model more robust even with few training samples. The presented method provides excellent results compared to more sophisticated but solely deep learning or traditional computer vision techniques as shown in “MapSeg” segmentation competition, where all other approaches were significantly outperformed. We further propose two additional approaches that improve the original method and set a new state-of-the-art result on the MapSeg dataset. The methods are further tested on an extended version of the Map Border dataset to show their robustness.
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
References
Aurelie, L., Jean, C.: Segmentation of historical maps without annotated data. In: The 6th International Workshop on Historical Document Imaging and Processing, pp. 19–24 (2021)
Baloun, J., Král, P., Lenc, L.: Chronseg: Novel dataset for segmentation of handwritten historical chronicles. In: Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, pp. 314–322. INSTICC, SciTePress (2021). https://doi.org/10.5220/0010317203140322
Baloun, J., Lenc, L., Král, P.: Robust grid detection in historical map images. In: 2022 IEEE International Conference on Image Processing (ICIP), pp. 1931–1935 (2022). https://doi.org/10.1109/ICIP46576.2022.9897721
Chazalon, J., et al.: ICDAR 2021 competition on historical map segmentation. In: Lladós, J., Lopresti, D., Uchida, S. (eds.) ICDAR 2021. LNCS, vol. 12824, pp. 693–707. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-86337-1_46
Chazalon, J., et al.: Icdar 2021 competition on historical map segmentation - dataset. online dataset (2021). https://doi.org/10.5281/zenodo.4817662
Chazalon, J., Edwin-Carlinet: icdar21-mapseg/icdar21-mapseg-eval: zenodo archival (2021). https://doi.org/10.5281/zenodo.4818400
Chen, Y., Carlinet, E., Chazalon, J., Mallet, C., Duménieu, B., Perret, J.: Combining deep learning and mathematical morphology for historical map segmentation. In: Lindblad, J., Malmberg, F., Sladoje, N. (eds.) DGMM 2021. LNCS, vol. 12708, pp. 79–92. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-76657-3_5
Chen, Y., Carlinet, E., Chazalon, J., Mallet, C., Duménieu, B., Perret, J.: Vectorization of historical maps using deep edge filtering and closed shape extraction. In: Lladós, J., Lopresti, D., Uchida, S. (eds.) ICDAR 2021. LNCS, vol. 12824, pp. 510–525. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-86337-1_34
Foroughi, F., Wang, J., Nemati, A., Chen, Z., Pei, H.: Mapsegnet: a fully automated model based on the encoder-decoder architecture for indoor map segmentation. IEEE Access 9, 101530–101542 (2021)
Garcia-Molsosa, A., Orengo, H.A., Lawrence, D., Philip, G., Hopper, K., Petrie, C.A.: Potential of deep learning segmentation for the extraction of archaeological features from historical map series. Archaeol. Prospect. 28(2), 187–199 (2021)
Lenc, L., Martínek, J., Baloun, J., Prantl, M., Král, P.: Historical map toponym extraction for efficient information retrieval. In: Uchida, S., Barney, E., Eglin, V. (eds.) DAS 2022. LNCS, vol. 13237, pp. 171–183. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-06555-2_12
Lenc, L., Prantl, M., Martínek, J., Král, P.: Border detection for seamless connection of historical cadastral maps. In: Barney Smith, E.H., Pal, U. (eds.) ICDAR 2021. LNCS, vol. 12916, pp. 43–58. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-86198-8_4
Li, Z., Chen, J.: Superpixel segmentation using linear spectral clustering. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1356–1363 (2015)
Liu, T., Miao, Q., Xu, P., Song, J., Quan, Y.: Color topographical map segmentation algorithm based on linear element features. Multimedia Tools Appl. 75(10), 5417–5438 (2016)
Liu, T., Miao, Q., Xu, P., Zhang, S.: Superpixel-based shallow convolutional neural network (SSCNN) for scanned topographic map segmentation. Remote Sens. 12(20), 3421 (2020)
Min, J., Kang, D., Cho, M.: Hypercorrelation squeeze for few-shot segmentation. CoRR abs/2104.01538 (2021). https://arxiv.org/abs/2104.01538
Nina, O., Morse, B., Barrett, W.: A recursive OTSU thresholding method for scanned document binarization. In: 2011 IEEE Workshop on Applications of Computer Vision (WACV), pp. 307–314. IEEE (2011)
Peterson, J.C., Battleday, R.M., Griffiths, T.L., Russakovsky, O.: Human uncertainty makes classification more robust. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), October 2019
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Acknowledgements
This work has been partly supported by the Grant No. SGS-2022-016 Advanced methods of data processing and analysis.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Baloun, J., Lenc, L., Král, P. (2023). FCN-Boosted Historical Map Segmentation with Little Training Data. In: Fink, G.A., Jain, R., Kise, K., Zanibbi, R. (eds) Document Analysis and Recognition - ICDAR 2023. ICDAR 2023. Lecture Notes in Computer Science, vol 14187. Springer, Cham. https://doi.org/10.1007/978-3-031-41676-7_30
Download citation
DOI: https://doi.org/10.1007/978-3-031-41676-7_30
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-41675-0
Online ISBN: 978-3-031-41676-7
eBook Packages: Computer ScienceComputer Science (R0)
