Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Deep Learning Models for Judging Business Card Image Rotation

명함 이미지 회전 판단을 위한 딥러닝 모델 비교

  • Ji-Hoon, Kyung (Department of Industrial & Management Engineering, Hannam University)
  • Received : 2022.11.26
  • Accepted : 2022.12.27
  • Published : 2023.01.31
Download PDF
⟨ Previous Next ⟩

Abstract

A smart business card printing system that automatically prints business cards requested by customers online is being activated. What matters is that the business card submitted by the customer to the system may be abnormal. This paper deals with the problem of determining whether the image of a business card has been abnormally rotated by adopting artificial intelligence technology. It is assumed that the business card rotates 0 degrees, 90 degrees, 180 degrees, and 270 degrees. Experiments were conducted by applying existing VGG, ResNet, and DenseNet artificial neural networks without designing special artificial neural networks, and they were able to distinguish image rotation with an accuracy of about 97%. DenseNet161 showed 97.9% accuracy and ResNet34 also showed 97.2% precision. This illustrates that if the problem is simple, it can produce sufficiently good results even if the neural network is not a complex one.

고객이 온라인으로 요청한 명함을 자동으로 명함을 인쇄하는 스마트 명함 인쇄 시스템이 활성화되고 있다. 이때, 문제는 고객이 시스템에 제출한 명함이 비정상일 수 있다는 것이다. 본 논문에서는 인공 지능 기술을 도입하여 명함의 이미지가 비정상적으로 회전됐는지 여부를 판정하는 문제를 다룬다. 명함은 0도, 90도, 180도, 270도 회전한다고 가정하였다. 특별한 인공신경망을 설계하지 않고 기존의 VGG, ResNet, DenseNet 인공신경망을 적용하여 실험하였는데 모든 신경망이 97% 정도의 정확도로 이미지 회전을 분별할 수 있었다. DenseNet161은 97.9%의 정확도를 보였고 ResNet34도 97.2%의 정밀도를 보였다. 이는 문제가 단순할 경우, 복잡한 인공신경망이 아니어도 충분히 좋은 결과를 낼 수 있음을 시사한다.

Keywords

Acknowledgement

This work was supported by 2021 Hannam University Research Fund.

References

  1. E. J. Park and B. H. Ha, "A Formal Framework for Analyzing Performance of Container Terminal Operations," Journal of Society for e-Business Studies, vol. 18, no. 2, pp. 191-203, May 2013. DOI: 10.7838/jsebs.2013.18.2.191.
  2. R. Seiger, L. Malburg, B. Weber, and R. Bergmann, "Integrating process management and event processing in smart factories: A systems architecture and use cases," Journal of Manufacturing systems, vol. 63, pp. 575-592, Apr. 2022. DOI: 10.1016/j.jmsy.2022.05.012.
  3. H. U. Park, "Trends in production and manufacturing technology related to smart factories," Information and Communications Magazine, vol. 33, no. 1, pp. 24-29, Dec. 2015.
  4. K. S. Ko, J. J. Huh, and J. I. Oh, "A Study on the Factors that Affect the Adoption of a Smart Factory - Focusing on the Comparison between Customers and Suppliers," Korea Business Review, vol. 25, no. 3, pp. 129-151, Aug. 2021. DOI: 10.17287/kbr.2021.25.3.129.
  5. D. Y. Son and K. K. Lee, "A Study on the Recognition of Face Based on CNN Algorithms," Korean Journal of Artificial Intelligence, vol. 5, no. 2, pp. 15-22, Dec. 2017. DOI: 10.24225/kjai.2017.5.2.15.
  6. Y. S. Kwon, D. J. Shin, and J. J. Kim, "A Study on Application Method of Contour Image Learning to improve the Accuracy of CNN by Data," The Journal of The Institute of Internet, Broadcasting and Communication (IIBC), vol. 22, no. 4, pp. 171-176, Aug. 2022. DOI: 10.7236/JIIBC.2022.22.4.171.
  7. J. W. Kim, H. Pyo, J. Ha, C. Lee, and J. Kim, "Various deep learning algorithms and applications," Communications of the Korean Institute of Information Scientists and Engineers, vol. 33, no. 8, pp. 25-31, Aug. 2015.
  8. B. M. Kim, "Trend of image classification technology based on deep learning," Korea Institute of Communication Sciences, vol. 35, no. 12, pp. 8-14, Nov. 2018.
  9. D. Lee, and et al., "CNN-based Image Rotation Correction Algorithm to Improve Image Recognition Rate," The Journal of The Institute of Internet, Broadcasting and Communication, vol. 20, no. 1, pp. 225-229, Feb. 2020.
  10. A. Das, S. Roy, U. Bhattacharya, and S. K. Parui, "Document Image Classification with Intra-Domain Transfer Learning and Stacked Generalization of Deep Convolutional Neural Networks," in Proceedings of 2018 24th International Conference on Pattern Recognition (ICPR), Beijing, China, pp 3180-3185, 2018. DOI: 10.1109/ICPR.2018.8545630.
  11. X. Deng. Y. Zhang, S. Yang, P. Tan, L. Chang, Y. Yuan, and H. Wang, "Joint Hand Detection and Rotation Estimation Using CNN," IEEE transactions on image processing, vol. 27, no. 1, pp. 1888-1900, Apr. 2018. DOI: 10.1109/TIP.2017.2779600.
  12. K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," in Proceedings of conference paper at ICLR 2015, San Diego: CA, USA, pp. 7-9, 2015.
  13. K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," in Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas: NV, USA, pp. 770-778, Jun. 2016.
  14. G. Huang, Z. Liu, L, and K. Q. Weinberger, "Densely Connected Convolutional Networks," in Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu: HI, USA, pp. 4700-4708, Jul. 2017.
  15. D. Brunet, E. R. Vrscay, and Z. Wang, "On the Mathematical Properties of the Structural Similarity Index," IEEE transactions on image processing, vol. 21, no. 4, pp. 1488-1499, April. 2012. https://doi.org/10.1109/TIP.2011.2173206