Journal of the Institute of Electronics and Information Engineers (전자공학회논문지)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지
DOI QR코드

DOI QR Code

Downscaling Forgery Detection using Pixel Value's Gradients of Digital Image

디지털 영상 픽셀값의 경사도를 이용한 Downscaling Forgery 검출

  • RHEE, Kang Hyeon (Chosun University, Dept. of Electronics Eng. / School of Design and Creative Eng.)
  • 이강현 (조선대학교 전자공학과 창의공학디자인융합학과)
  • Received : 2015.11.18
  • Accepted : 2016.01.17
  • Published : 2016.02.25
Download PDF
⟨ Previous Next ⟩

Abstract

The used digital images in the smart device and small displayer has been a downscaled image. In this paper, the detection of the downscaling image forgery is proposed using the feature vector according to the pixel value's gradients. In the proposed algorithm, AR (Autoregressive) coefficients are computed from pixel value's gradients of the image. These coefficients as the feature vectors are used in the learning of a SVM (Support Vector Machine) classification for the downscaling image forgery detector. On the performance of the proposed algorithm, it is excellent at the downscaling 90% image forgery compare to MFR (Median Filter Residual) scheme that had the same 10-Dim. feature vectors and 686-Dim. SPAM (Subtractive Pixel Adjacency Matrix) scheme. In averaging filtering ($3{\times}3$) and median filtering ($3{\times}3$) images, it has a higher detection ratio. Especially, the measured performances of all items in averaging and median filtering ($3{\times}3$), AUC (Area Under Curve) by the sensitivity and 1-specificity is approached to 1. Thus, it is confirmed that the grade evaluation of the proposed algorithm is 'Excellent (A)'.

스마트 기기와 소형 디스플레이에 사용되는 디지털 영상은 다운스케일링 (Downscaling)된 영상이 사용된다. 본 논문에서는 영상 픽셀값의 경사도에 따른 특징벡터를 이용한 다운스케일링 포저리 (Forgery) 영상 검출 알고리즘을 제안한다. 제안된 알고리즘에서, 원영상의 픽셀값 경사도로부터 자기회귀 (AR: Autoregressive) 계수를 계산한다. 이는 다운스케일링 포저리 영상 검출기의 SVM (Support Vector Machine) 분류를 위한 학습에 사용된다. 제안된 다운스케일링 검출 알고리즘은 동일 10-Dim. 특징벡터의 MFR (Median Filter Residual) 스킴과 686-Dim.의 SPAM (Subtractive Pixel Adjacency Matrix) 스킴과 비교하여 다운스케일링 90% 영상 포저리에서 성능이 우수하며, 평균필터링 ($3{\times}3$) 영상과 미디언필터링 ($3{\times}3$) 영상에서 높은 검출율을 보여 주었다. 특히, 평균필터링과 미디언필터링 영상에서는 성능평가 전체 항목에서 민감도 (Sensitivity; TP: True Positive rate)와 1-특이도 (1-Specificity; FP: False Positive rate)의 AUC (Area Under Curve)가 모두 1에 수렴하여 'Excellent (A)' 등급임을 확인하였다.

Keywords

References

  1. Kang Hyeon RHEE, "Median Filtering Detection using Latent Growth Modeling," THE INSTITUTE OF ELECTRONICS AND INFORMATION ENGINEERS, Journal of The Institute of Electronics and Information Engineers, Vol. 52, No. 1, pp. 61-68, 2015.1. https://doi.org/10.5573/ieie.2015.52.1.061
  2. Kang Hyeon RHEE, "Image Forensic Decision Algorithm using Edge Energy Information of Forgery Image," THE INSTITUTE OF ELECTRONICS AND INFORMATION ENGINEERS, Journal of The Institute of Electronics and Information Engineers, Vol. 51, No. 3, pp. 75-81, 2014.3. https://doi.org/10.5573/ieie.2014.51.3.075
  3. Kang Hyeon RHEE, "Forensic Decision of Median Filtering by Pixel Value's Gradients of Digital Image," THE INSTITUTE OF ELECTRONICS AND INFORMATION ENGINEERS, Journal of The Institute of Electronics and Information Engineers, Vol. 52, No. 6, pp. 79-84, 2015.6. https://doi.org/10.5573/ieie.2015.52.6.079
  4. Kang Hyeon RHEE, "Forensic Decision of Median Filtering Image Using a Coefficient of Variation of Fourier Transform," THE INSTITUTE OF ELECTRONICS AND INFORMATION ENGINEERS, Journal of The Institute of Electronics and Information Engineers, Vol. 52, No. 8, 2015.8.
  5. Xiangui Kang, Matthew C. Stamm, Anjie Peng, and K. J. Ray Liu, "Robust Median Filtering Forensics Using an Autoregressive Model," IEEE Trans. on Information Forensics and Security, vol. 8, no. 9, pp. 1456-1468, Sept. 2013. https://doi.org/10.1109/TIFS.2013.2273394
  6. Tomas Pevny, "Steganalysis by Subtractive Pixel Adjacency Matrix," Information Forensics and Security, IEEE Transactions on, Vol. 5, pp. 215-224, 2010. https://doi.org/10.1109/TIFS.2010.2045842
  7. Kang Hyeon RHEE, "Median Filtering Detection Using Variation of Neighboring Line Pairs for Image Forensic," The 5th IEEE ICCE-Berlin Sept. 6-9, 2015, pp. 103-107, Sept. 2015
  8. http://homepages.lboro.ac.uk/-cogs/datasets/ucid/ucid.html (2015.4.1)