Recognition of cadastral coordinates using convolutional recurrent neural networks

Cover Page

Cite item

Full Text

Abstract

The article examines the use of convolutional recurrent neural networks (CRNN) for recognizing images of cadastral coordinates of objects on scanned documents of the «Roskadastr» PLC. The combined CRNN architecture, combining convolutional neural networks (CNN) and recurrent neural networks (RNN), allows you to take advantage of each of them for image processing and recognition of continuous digital sequences contained in them. During experimental studies, images consisting of a given number of digits were generated, and a CRNN model was built and studied. The formation of images of digital sequences consisted of preprocessing and concatenation of images of the digits forming them from one's own data set. Analysis of the values of the loss function and Accuracy, Character Error Rate (CER), and Word Error Rate (WER) metrics showed that the use of the proposed CRNN model makes it possible to achieve high accuracy in recognizing cadastral coordinates in their scanned images.

About the authors

Igor Victorovich Vinokurov

Financial University under the Government of the Russian Federation

Author for correspondence.
Email: igvvinokurov@fa.ru
ORCID iD: 0000-0001-8697-1032
Candidate of Technical Sciences (PhD), Associate Professor at the Financial University under the Government of the Russian Federation. Research interests: information systems, information technologies, data processing technologies

References

  1. Shi B., Bai X., Yao C.. “An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition”, IEEE Transactions on Pattern Analysis and Machine Intelligence, 39:11 (2017), pp. 2298–2304.
  2. Hochreiter S., Schmidhuber J.. “Long short-term memory”, Neural Computation, 9:8 (1997), pp. 1735–1780.
  3. Chung J., Gulcehre C., Cho K., Bengio Y.. “Gated feedback recurrent neural networks”, Proceedings of Machine Learning Research, 37 (2015), pp. 2067–2075.
  4. Винокуров И. В.. «Использование свёрточной нейронной сети для распознавания элементов текста на отсканированных изображениях плохого качества», Программные системы: теория и приложения, 13:3(54) (2022), с. 29–43.
  5. Винокуров И. В.. «Распознавание табличной информации с использованием свёрточных нейронных сетей», Программные системы: теория и приложения, 14:1(56) (2023), с. 3–30.
  6. Винокуров И. В.. «Распознавание цифровых последовательностей с использованием свёрточных нейронных сетей», Программные системы: теория и приложения, 14:3(58) (2023), с. 3–36.
  7. He P., Huang W., Qiao Y., Change Loy C., Tang X.. “Reading scene text in deep convolutional sequences”, AAAI'16: Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (Phoenix, Arizona, USA, February 12–17, 2016), Proceedings of the AAAI Conference on Artificial Intelligence, 30 (2016), pp. 3501–3508.
  8. Shi B., Wang X., Lv P., Yao C., Bai X.. “Robust scene text recognition with automatic rectification”, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (Las Vegas, NV, USA, June 27–30, 2016), 2016, pp. 4168–4176.
  9. Yin F., Wu Y. -C., Zhang X. -Y., Liu C. -L.. Scene text recognition with sliding convolutional character models, 2017, 10 pp.
  10. Nirmalasari D. A., Suciati N., Navastara D. A.. “Handwritten text recognition using fully convolutional network”, IOP Conference Series: Materials Science and Engineering, 1077:1 (2021), 012030, 9 pp.
  11. Liu X., Deng Y., Sun Y., Zhou Y.. “Multi-digit recognition with convolutional neural network and long short-term memory”, 2018 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) (Huangshan, China, July 28–30, 2018), IEEE, 2018, pp. 1187–1192.
  12. Madakannu A., Selvaraj A.. “DIGI-Net: a deep convolutional neural network for multi-format digit recognition”, Neural Computing and Applications, 32 (2020), pp. 11373–11383.
  13. Zou L., He Z., Wang K., Wu Z., Wang Y., Zhang G., Wang X.. “Text recognition model based on multi-scale fusion CRNN”, Sensors, 32:16 (2023), 7034, 18 pp.
  14. Agrawal V., Jagtap J.. Convolutional vision transformer for handwritten digit recognition, Research Square, 2022, 11 pp.
  15. Cheng L., Khalitov R., Yu T., Yang Z.. Classification of long sequential data using circular dilated convolutional neural networks, 2022, 16 pp.
  16. Bhat R. S.. Text recognition with CRNN-CTC network, W&B Fully Connected, 2022 URL https://wandb.ai/authors/text-recognition-crnn-ctc/reports/Text-Recognition-With-CRNN-CTC-Network–VmlldzoxNTI5NDI.
  17. Khamekhem S., Sourour A., Kessentini Y.. “Domain and writer adaptation of offline Arabic handwriting recognition using deep neural networks”, Neural Computing and Applications, 34 (2022), pp. 2055–2071.

Supplementary files

Supplementary Files
Action
1. JATS XML
Download


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Согласие на обработку персональных данных

 

Используя сайт https://journals.rcsi.science, я (далее – «Пользователь» или «Субъект персональных данных») даю согласие на обработку персональных данных на этом сайте (текст Согласия) и на обработку персональных данных с помощью сервиса «Яндекс.Метрика» (текст Согласия).