Journal of Advanced Navigation Technology (한국항행학회논문지)

The Korean Navigation Institute (한국항행학회)
권호 표지
DOI QR코드

DOI QR Code

Code Embedding and Detection for Camera Identifiable Photos

카메라 식별 가능 사진을 위한 코드탑재 및 검출

  • Bae, Joon-Hyeon (Dept. of Computer Engineering, Kumoh National Institute of Technology) ;
  • Hong, Won-Kee (Dept. of Multimedia Engineering, Daegu University) ;
  • Ko, Jaepil (Dept. of Computer Engineering, Kumoh National Institute of Technology)
  • 배준현 (금오공과대학교 컴퓨터공학과) ;
  • 홍원기 (대구대학교 멀티미디어공학전공) ;
  • 고재필 (금오공과대학교 컴퓨터공학과)
  • Received : 2021.03.16
  • Accepted : 2021.04.28
  • Published : 2021.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, applications that acquire additional information by taking a quick response (QR) code photo with a smart-phone camera is widely used. The QR codes can contain a lot of information, but in order to recognize a QR code, we need to take a photo so that all four corners of the code are covered. To overcome this limitation, we propose a method to embed a sequence of code bits to a digital image and detect from the captured image of its photos. The proposed scheme embed the code in the frequency domain so that the code is not lost despite severe image degradation issued during the print-camera process. The conventional frequency-domain approaches usually embed only 1~2 bits, however we designed to be able to contain 16 bits. In particular, we additionally embed alignment bits as well as code bits to facilitate the decoding process. In our experiments, we achieved an average accuracy of 95.2% for images partially captured under random rotation. This corresponds to only 0.76 bits in terms of the average number of error bits.

최근 스마트폰 카메라로 QR (quick response) 코드사진을 촬영하여 추가 정보를 얻는 응용이 널리 사용되고 있다. QR 코드는 대용량 정보를 탑재할 수 있으나, 이를 인식하기 위해서는 반드시 코드의 네 모서리가 모두 포함되도록 촬영해야 한다. 본 논문에서는 이러한 제약을 극복할 수 있도록 디지털 영상에 코드 비트열을 탑재하고 검출하는 기법을 제안한다. 제안하는 기법은 프린트-캡처과정에서 발행하는 심한 영상 훼손에도 불구하고 탑재된 코드가 손실되지 않도록 주파수 도메인에 코드를 탑재한다. 기존의 주파수 도메인 탑재 방법은 탑재 비트수가 1~2비트 수준이나 본 논문에서는 16비트를 탑재할 수 있도록 설계하였다. 특히, 코드 비트열 외에 정렬 비트를 추가로 탑재하여 복호화 과정을 용이하도록 하였다. 연구실에서 수집한 영상에 대한 실험에서 임의 회전 하에서 일부만 촬영된 영상에 대해서도 평균 인식률 95.2%를 달성하였다. 이것은 평균 오류 비트의 수로 따지면 0.76개에 불과하다.

Keywords

Acknowledgement

이 연구는 금오공과대학교 학술연구비로 지원되었음 (2018104083).

References

  1. M. Hirokawa, and J. Iijima, "A study on digital watermarking usage in the mobile marketing field: cases in Japan," in Proceedings of IEEE Symposium on Logistics and Industrial Informatics, Linz: Austria, pp. 1-6, Sept. 2009.
  2. M. Tancik, B. Mildenhall., and R. Ng., "Stegastamp: invisible hyperlinks in physical photographs", in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Seattle: WA, pp. 2117-2126, June 2020.
  3. D. He, and Q. Sun, "A practical print-scan resilient watermarking scheme," in Proceedings of IEEE Conference on Image Processing, Genova: Italy, pp. 257, Sept. 2005.
  4. T. Nakamura, A. Katayama, M. Yamamuro, and N. Sonehara, "Fast watermark detection scheme for camera-equipped cellular phone," in Proceedings of ACM Conference on Mobile and Ubiquitous Multimedia, New York: NY, pp. 101-108, Oct. 2004.
  5. A. Pramil, A. Keskinarkaus, and T. Seppanen, "Toward an interactive poster using digital watermarking and a mobile phone camera," Signal, Image and Video Processing, Vol. 6, No. 2, pp. 211-222, 2012. https://doi.org/10.1007/s11760-011-0211-2
  6. A. Poljicak, L. Mandic, and D. Agic, "Discrete fourier transform-based watermarking method with an optimal implementation radius," Journal of Electronic Imaging, Vol. 20, No. 3, pp. 033008-1-033008-8, 2011. https://doi.org/10.1117/1.3609010
  7. R. C. Gonzalez, and R. E. Woods, Digital Image Processing, Addison-Wesley Pub., ch. 3, pp. 120-128, 1992.
  8. K. Zhang., A. Cuesta-Infane, L. Xu., and K. Veermachaneni., "SteganoGAN: high capacity image steganography with GANs," ArXiv, 2019. Available: https://arxiv.org/abs/1901.03892
  9. J. Liu, Y. Ke, and X. Yang, "Recent advances of image steganography with generative adversarial networks," IEEE Access, Vol. 8, pp. 60575-60579, 2020. https://doi.org/10.1109/access.2020.2983175