한국지능시스템학회논문지 (Journal of the Korean Institute of Intelligent Systems)

  • 제26권5호
  • /
  • Pages.375-382
  • /
  • 2016
  • /
  • 1976-9172(pISSN)
  • /
  • 2288-2324(eISSN)
한국지능시스템학회 (Korean Institute of Intelligent Systems)
권호 표지
DOI QR코드

DOI QR Code

다차원 인접화소 간 명암차의 극좌표 기반 비선형 양자화 히스토그램에 의한 서명인식

Signatures Verification by Using Nonlinear Quantization Histogram Based on Polar Coordinate of Multidimensional Adjacent Pixel Intensity Difference

  • 조용현 (대구가톨릭대학교 IT공학부)
  • Cho, Yong-Hyun (School of Information Technology, Catholic University of Daegu)
  • 투고 : 2016.09.29
  • 심사 : 2016.10.10
  • 발행 : 2016.10.25
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 다차원 인접화소 간 명암차 기반 극좌표의 비선형 양자화 히스토그램을 이용한 서명인식을 제안한다. 다차원 인접화소 간 명암차는 기준화소를 중심으로 횡방향, 종방향, 대각 방향, 역대각 방향 각각의 이웃화소 간 명암차이고, 극좌표는 횡과 종 방향 및 대각과 비대각 방향 각각의 직교좌표로부터 변환된 좌표이며, 비선형 양자화 히스토그램은 반복계산 기법인 Lloyd 알고리즘에 의해 극좌표 값을 비균일 양자화한 히스토그램이다. 여기서 4방향 명암차의 극좌표 히스토그램은 대응하는 화소간의 상관성을 좀 더 많이 고려할 뿐만 아니라 히스토그램의 수를 감소시켜 계산부하를 줄이기 위함이다. 또한 비선형 양자화는 화소간의 명암변화의 속성을 더욱 더 잘 반영할 뿐만 아니라 저차원의 히스토그램 레벨을 얻기 위함이다. 제안된 기법을 256*256 픽셀의 90개(3인*30개) 서명들을 대상으로 city-block거리, Euclidean 거리, 순서값, 그리고 정규상호상관계수 각각의 정합척도에 기반 한 실험결과, 선형 양자화 기반 히스토그램에 비해 우수한 인식성능을 가지며, Euclidean 거리가 가장 우수한 정합척도임을 확인하였다.

In this paper, we presents a signatures verification by using the nonlinear quantization histogram of polar coordinate based on multi-dimensional adjacent pixel intensity difference. The multi-dimensional adjacent pixel intensity difference is calculated from an intensity difference between a pair of pixels in a horizontal, vertical, diagonal, and opposite diagonal directions centering around the reference pixel. The polar coordinate is converted from the rectangular coordinate by making a pair of horizontal and vertical difference, and diagonal and opposite diagonal difference, respectively. The nonlinear quantization histogram is also calculated from nonuniformly quantizing the polar coordinate value by using the Lloyd algorithm, which is the recursive method. The polar coordinate histogram of 4-directional intensity difference is applied not only for more considering the corelation between pixels but also for reducing the calculation load by decreasing the number of histogram. The nonlinear quantization is also applied not only to still more reflect an attribute of intensity variations between pixels but also to obtain the low level histogram. The proposed method has been applied to verified 90(3 persons * 30 signatures/person) images of 256*256 pixels based on a matching measures of city-block, Euclidean, ordinal value, and normalized cross-correlation coefficient. The experimental results show that the proposed method has a superior to the linear quantization histogram, and Euclidean distance is also the optimal matching measure.

키워드

참고문헌

  1. D. Impedovo and G. Pirlo, "Automatic Signature Verification : The State of the Art," IEEE. Trans. on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Vol. 38, No. 5, pp. 609-928, Sept. 2008. https://doi.org/10.1109/TSMCC.2008.923866
  2. A. A Kholmatov, "Biometric Identity Verification Using On-line & Off-line Signature Verification", Master of Science Thesis, Sabanci University, Spring 2003.
  3. B. A. Kovari, Models and Algorithms in Off-line, Featurebased, Handwritten Signature Verification, Budapest Univ. of Tech. and Economics, PhD Thesis, 2013.
  4. M. Vasta, R. Singh, P. Mitra, and A. Noore, "Signature Verification Using Static and Dynamic Features," ICONIP 2004, LNCS 3316, pp. 350-355, Nov. 2004.
  5. A. P. Shanker and A. N. Rajagopalan, "Off-line Signature Verification Using DTW," Pattern Recognition Letters, No. 28, pp. 1407-1414, Sept. 2007. https://doi.org/10.1016/j.patrec.2007.02.016
  6. Y. H. Cho, "An Efficient Signature Recognition Based on Histogram Using Statistical Characteristics," Journal of Korean Institute of Intelligent Systems, Vol. 20, No. 5, pp. 701-709, Oct. 2010. https://doi.org/10.5391/JKIIS.2010.20.5.701
  7. S. Lloyd, "Least Squares Quantization in PCM," IEEE Trans. on Information Theory, Vol. 28, No. 2, pp. 129-137, Mar. 1982. https://doi.org/10.1109/TIT.1982.1056489
  8. M. Mayer, Quantization of Images and Lloyd'Algorithm, Vienna Univ. of Tech., Bachelor Thesis, Sept. 2010.
  9. S. H. Cha and S. N. Srihari, "On Measuring the Distance between Histograms," Pattern Recognition Letters, No. 35, pp. 1355-1370, Sept. 2002. https://doi.org/10.1016/S0031-3203(01)00118-2
  10. F. Serratosa and A. Sanfeliu, "A Fast Distance Between Histograms," 10th Iberoamerican Congress on Pattern Recognition, LNCS 3316, pp. 1027-1035, Nov. 2005.
  11. F. Lee, K. Kotani, Q. Chen, and T. Ohmi, "Face Recognition Algorithm Using Adjacent Pixel Intensity Difference Quantization Histogram," International Journal of Computer Science and Network Security, Vol. 9, No. 8, pp. 147-154, Aug. 2009.
  12. Y. H. Cho, "Image Recognition Based on Nonlinear Equalization and Multidimensional Intensity Variation," Journal of Korean Institute of Intelligent Systems, Vol. 24, No. 5, pp. 504-511, Oct. 2014. https://doi.org/10.5391/JKIIS.2014.24.5.504
  13. Y. H. Cho, "Image A Texture Classification Based on LBP by Using Intensity Differences between Pixels Variation," Journal of Korean Institute of Intelligent Systems, Vol. 25, No. 5, pp. 483-488, Oct. 2015. https://doi.org/10.5391/JKIIS.2015.25.5.483