한국산업정보학회논문지 (Journal of Korea Society of Industrial Information Systems)

한국산업정보학회 (Korea Society of Industrial Information Systems)
권호 표지
DOI QR코드

DOI QR Code

딥러닝에 의한 한글 필기체 교정 어플 구현

An Implementation of Hangul Handwriting Correction Application Based on Deep Learning

  • 이재형 (국립한밭대학교 정보통신공학과) ;
  • 조민영 (국립한밭대학교 정보통신공학과) ;
  • 김진수 (국립한밭대학교 지능미디어공학과)
  • Jae-Hyeong Lee ;
  • Min-Young Cho ;
  • Jin-soo Kim
  • 투고 : 2024.04.16
  • 심사 : 2024.05.10
  • 발행 : 2024.06.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

현재 디지털 기기의 확산과 함께 일상에서 손으로 쓰는 글씨의 비중은 점점 줄어들고 있다. 키보드와 터치스크린의 활용도 증가에 따라 한글 필기체의 품질 저하는 어린 학생부터 성인까지 넓은 범위의 한글 문서에서 관찰되고 있다. 그러나 한글 필기체는 여전히 개인적인 고유한 특징을 포함하면서 가독성을 제공하는 많은 문서 작성에 필요하다. 이를 위해 본 논문에서는 손으로 쓴 한글 필기체의 품질을 개선하고, 교정하기 위한 목적의 어플 구현을 목적으로 한다. 제안된 어플은 CRAFT(Character-Region Awareness For Text Detection) 모델을 사용하여 필기체 영역을 검출하고, 딥러닝으로서 VGG-Feature-Extraction 모델을 사용하여 필기체의 특징을 학습한다. 이때 사용자가 작성한 한글 필기체의 음절 단위로 신뢰도를 인식률로 제시하고, 또한, 후보 폰트들중에서 가장 유사한 글자체를 추천하도록 구현한다. 다양한 실험을 통해 제안한 어플은 기존의 상용화된 문자 인식 소프트웨어와 비교할만한 우수한 인식률을 제공함을 확인할 수 있다.

Currently, with the proliferation of digital devices, the significance of handwritten texts in daily lives is gradually diminishing. As the use of keyboards and touch screens increase, a decline in Korean handwriting quality is being observed across a broad spectrum of Korean documents, from young students to adults. However, Korean handwriting still remains necessary for many documentations, as it retains individual unique features while ensuring readability. To this end, this paper aims to implement an application designed to improve and correct the quality of handwritten Korean script The implemented application utilizes the CRAFT (Character-Region Awareness For Text Detection) model for handwriting area detection and employs the VGG-Feature-Extraction as a deep learning model for learning features of the handwritten script. Simultaneously, the application presents the user's handwritten Korean script's reliability on a syllable-by-syllable basis as a recognition rate and also suggests the most similar fonts among candidate fonts. Furthermore, through various experiments, it can be confirmed that the proposed application provides an excellent recognition rate comparable to conventional commercial character recognition OCR systems.

키워드

참고문헌

  1. AI Hub (2023), https://www.aihub.or.kr/ (Accessed on March 12th, 2023)
  2. Baek, J., Seo, J., Jung, S. and Jeong, D. (2022), A Study on Hangul Handwriting Generation and Classification Mode for Intelligent OCR System, Journal of the Institute of Convergence Signal Processing, 23(4), 222-227.
  3. Baek, Y., Lee, B., Han, D., Yun, S. and Lee, H. (2019), Character Region Awareness for Text Detection, Computer Vision and Pattern Recognition, https://doi.org/10.48550/arXiv.1904.01941
  4. Belval, E. (2022), Text Recognition Data Generator, https://github.com/Belval/TextReco gnitionDataGenerator (Accessed on Oct. 10th, 2023)
  5. Clova AI Research, NAVER Corp. (2019), CRAFT: Character-Region Awareness For Text Detection, https://github.com/clovaai/CRAFT-pytorch (Accessed on May 12th, 2023)
  6. Deep Text Recognition Benchmark (2019), https://github.com/clovaai/deep-text-recognition-benchmark (Accessed on Oct. 10th, 2023)
  7. Imam, N., Vassilakis, V. and Kolovos, D. (2022), OCR Post-correction for Detecting Adversarial Text Images, Journal of Information Security and Applications, https://doi.org/10.1016/j.jisa.2022.103170
  8. Jeong, H. and Shin, C. (2023), Deep-Learning-based Smartphone Application for Automatic Recognition of Ingredients on Curved Containers, Journal of Korea Society of Industrial Information Systems, 28(6), 29-43.
  9. Keipour, A., Eshghi, M., Ghadikolaei, S., Mohammadi, N. and Ensafi, S. (2022), Omnifont Persian OCR System Using Primitives, Computer Vision and Pattern Recognition, https://doi.org/10.48550/arXiv.2202.06371
  10. Kwon, H. and Kang, H. (2022), Transformer Network for Container's BIC-code Recognition J ournal of Korea Society of Industrial Information Systems, 27(1), 19-26.
  11. Kwon, S. (2023), Structure Recognition Method of Invoice Document Image for Document Processing Automation, Journal of Korea Society of Industrial Information Systems, 28(2), 11-19.
  12. Lee, J., Cho, M. and Kim, J. (2023), Implementation of Handwriting Correction Application using Deep Learning, Proceedings of Spring Conference of Korea Society of Industrial Information Systems, June 1-3, Pusan, Korea.
  13. OpenCV, (2023). Open Source Computer Vision Library, https://opencv.org/(Accessed on May 12th, 2023)
  14. Shi, B., Bai, X. and Yao, C. (2015), 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), 2298-2304.
  15. Simonyan, K. and Zisserman, A. (2014), Very Deep Convolutional Networks for Large-Scale Image Recognition, Computer Vision and Pattern Recognition, https://doi.org/10.48550/arXiv.1409.1556
  16. Xiao, L., Ouayang, H. and Fan, C. (2019), An Improved Otsu Method for Threshold Segmentation Based on Set Mapping and Trapezoid Region Intercept Histogram, https://doi.org/10.1016/j.ijleo.2019.163106
  17. Zanwar, S., Shinede, U., Narote, A. and Narote, S. (2021), Hybrid Optimization and Effectual Classification for High Recognitions in OCR Systems, Journal of the Institution of Engineers(India) , 102(5), 966-977.