DCT기반 워터마크를 통한 위조 홍채영상 판별
Discrimination of Fake Iris Image by DCT-Based Watermark
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
생체인식 기술이 충분히 발전하여, 홍채, 지문 등의 생체정보를 사용하여 신원확인 하는 인증시스템이 다양한 분야에 적용되고 있다. 적용 분야가 증가하면서 생체정보를 위조하고자 하는 시도 또한 증가하고 있다. 본 논문에서는 눈매영역에 워터마크를 삽입하여 위조 홍채영상을 판별하는 방법을 제안한다. 제안하는 방법은 먼저 홍채영상에서 동공과 동공의 중심 그리고 눈매영역을 검출하고, 검출된 눈매영역을 8x8블록 단위로 DCT를 수행한 후, AC 영역에 가중치로 보정된 워터마크를 삽입 및 추출하여 위조 여부를 판별한다. 제안하는 방법에 대한 성능평가를 위하여, 본 논문은 CASIA Irisimage Database ver 4.0에서 제공하는 홍채영상들을 사용하며, PSNR(Peak Signal-to-noise ratio) 값으로 열화정도를 측정하고, NC(Normalized Correlation) 값을 사용하여 위조 여부를 판별한다.
Abstract
With biometric technology sufficiently advanced, authentication systems that identify a person using biometric information such as iris and fingerprint are being applied to various fields. While the applicable fields are increasing, but also the trials to generate fake biometric images are growing up. This paper proposes a technique discriminating fake iris images by inserting a watermark image on eyelid region. The proposed technique detects pupil and eyelid region on an iris image, performs DCT in 8x8 block of the detected eyelid region, and then inserts and extracts a watermark corrected by weight in the AC area to determine whether fake or not. In order to evaluate the performance of the proposed technique, this paper uses iris images provided in CASIA Irisimage Database ver 4.0, measures the deterioration level with PSNR, and calculates NC(Normalized Correlation) values to determine whether fake or not.
참고문헌 (16)
- A. Jain, "Handbook of Biometrics", Springer
- John Daugman, "Probing the uniqueness and random ness of iriscodes: results from 200 billion iris pair comparisons", Proc. IEEE, 94(11), pp.1927~1935 (2006)
- Younghwan Kim, "A motion and similarity-based fake detection method for biometric face recognition systems", IEEE Transac tions on Consumer Electronics, 57(2) (2011)
- Jonathan Connell, "Fake Iris Detection Using Structured Light", Acoustics, Speech and Signal Processing (ICASSP), pp.8692~8696 (2013)
- Ronaldo Martins da Costa, "Dyna mic Features for Iris Recognition", IEEE Trans. on System, Man, and Cy
- Eui Chul Lee, "Fake Iris Detection by Using Purkinje Image", ICB 2006:Advances in Biometrics, pp.397~403 (2006)
- Xiaofu, "A New Fake Iris Detecti on Method", ICB 2009: Advances in Biometrics, pp.1132~1139 (2009)
- I. J. Cox, "Digital Waterma rking" (2002), Academic Press
- V. M. Potdar, "A survey of digita l image watermarking techniques", IEEE Intl. Conf. Industrial Informatics, pp.709~716 (2005)
- P. Meerwald, "Survey of wavelet-domain watermarking algorithms", Proc. SPIE, Electronic Ima ging, Security and Watermarking of Multimedia Contents III, 4314, pp.505~516 (2001)
- C-T. Hsu, "Hidden digital watermarks in images", IEEE Trans. Image Processing, 8(1), pp.58~68 (1999)
- Shinfeng D. Lin, "Improving the robustness of DCT-based image watermar king against JPEG compression", International Conference on Consumer Electronics, pp.343~344 (2005)
- I. J. Cox, "Secure spread spectrum watermarking for multime dia", IEEE Trans. Image Processing, 6(12), pp.1673~1687 (1997)
- Mohammed A. M. Abdullah, "Securing Iris Images with a Robust Watermarking Algorithm based on Discrete Cosine Transform", Proceedings of the 10th International Conference on Computer Vision Theory and Applications, pp.108~114 (2015)
- 김만기, "워터마크를 이용한 위조 홍채 영상 식별", 한국컴퓨터정보학회논문지, 23(4), pp.33~39 (2018), 한국컴퓨터정보학회[DOI]
- HK Yuen, "Compara tive study of Hough Transform methods for circle finding", Image and Vision Computing, 8(1), pp.71~77 (1990)