Automatic Depth Value Recognition on Pilot Charts Using Deep Learning Methods

information from pilot charts. The relevance of this task is driven by the need to automate the processing of large volumes of cartographic data to create depth maps suitable for mathematical modelling of hydrodynamic and hydrobiological processes. The objective of this work is to develop the software tool LocMap, designed for the automatic detection and 52     recognition of depth values represented as numbers on pilot chart images.

Materials and Methods. The study employs deep learning methods, including convolutional neural networks (ResNet) for feature extraction, the Differentiable Binarization (DB) algorithm for text detection, and the Scene Text Recognition with a Single Visual Model (SVTR) architecture for text recognition.

Results. The developed software allows users to upload pilot chart images, perform preprocessing, detect and recognize depth values, highlight them in the image, and save the results in a text file. Testing results demonstrated that the system ensures high accuracy in recognizing depth values on pilot charts.

Discussion and Conclusion. The obtained results highlight the practical significance of the developed solution for automating the processing of pilot charts.

1. Lyashenko T.V., Chistyakov A.E., Nikitina A.V. Modelling the process of phosphate pollution spread in the aquatic ecosystem. Computational Mathematics and Information Technologies. 2023;6(4):47–53. (In Russ.) https://doi.org/10.23947/2587-8999-2023-7-4-47-53

2. Belova Yu.V., Filina A.A., Chistyakov A.E. Forecasting the dynamics of summer phytoplankton species based on satellite data assimilation methods. Computational Mathematics and Information Technologies. 2024;8(4):27–34. (In Russ.) https://doi.org/10.23947/2587-8999-2024-8-4-27-34

3. Belova Yu.V., Razveeva I.F., Rakhimbaeva E.O. Development of an Algorithm for Semantic Segmentation of Earth Remote Sensing Data to Determine Phytoplankton Populations. Advanced Engineering Research (Rostov-on-Don). 2024;24(3):283–292. https://doi.org/10.23947/2687-1653-2024-24-3-283-292

4. Cubuk E. D., Zoph B., Mané D., Vasudevan V. and Le Q. V., Autoaugment: Learning augmentation strategies from data[C]. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA, 2019. P. 113–123. https://doi.org/10.1109/CVPR.2019.00020

5. Minghui Liao, Zhaoyi Wan, Cong Yao, Chen Kai, Xiang Bai. Real-Time Scene Text Detection with Differentiable Binarization. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2020;34(07):11474–11481. https://doi.org/10.1609/aaai.v34i07.6812

6. Sheeba M.C., Christopher Ch. S. Adaptive deep residual network for image denoising across multiple noise levels in medical, nature, and satellite images. Ain Shams Engineering Journal. 2025;16(1):103188. https://doi.org/10.1016/j.asej.2024.103188

7. Bradley D., Roth G. Adaptive thresholding using the integral image. Journal of graphics tools. 2007;12(2):13–21. https://doi.org/10.1080/2151237X.2007.10129236

8. Tensmeyer C., Martinez T. Document image binarization with fully convolutional neural networks. In: 14th IAPR International Conference on Document Analysis and Recognition (ICDAR). 2017;1:99–104. https://doi.org/10.1109/ICDAR.2017.25

9. Minghui Liao, Baoguang Shi, Xiang Bai, Xinggang Wang, Wenyu Liu. Textboxes: A fast text detector with a single deep neural network. In: Proceedings of the AAAI Conference on Artificial Intelligence. 2017;31(1). https://doi.org/10.1609/aaai.v31i1.11196

10. Yongkun Du, Zhineng Chen, Caiyan Jia, et al. SVTR: Scene Text Recognition with a Single Visual Model. International Joint Conference on Artificial Intelligence (2022). https://doi.org/10.48550/arXiv.2205.00159

11. Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. Deep Residual Learning for Image Recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016. P. 770–778. https://doi.org/10.1109/CVPR.2016.90

12. Paschalis Tsirtsakis, Georgios Zacharis, George S. Maraslidis, George F. Fragulis. Deep learning for object recognition: A comprehensive review of models and algorithms. International Journal of Cognitive Computing in Engineering. 2025;6:298–312. https://doi.org/10.1016/j.ijcce.2025.01.004

13. Miao Tian, Kai Ma, Zhihao Liu, Qinjun Qiu, Yongjian Tan, Zhong Xie. Recognition of geological legends on a geological profile via an improved deep learning method with augmented data using transfer learning strategies. Ore Geology Reviews. 2023;153:105270. https://doi.org/10.1016/j.oregeorev.2022.105270

14. Sukhinov A.I., Belova Y.V., Filina A.A. Parallel implementation of substance transport problems for restoration the salinity field based on schemes of high order of accuracy. CEUR Workshop Proceedings (2019), 2500.

15. Nikitina A., Belova Y., Atayan A. Mathematical modeling of the distribution of nutrients and the dynamics of phytoplankton populations in the Azov Sea, taking into account the influence of salinity and temperature. AIP Conference Proceedings. 2019; 2188(1): 050027. https://doi.org/10.1063/1.5138454