Improved Run Length Based Detection of Digital Image Splicing

详细信息    查看全文

• 作者：Zhongwei He (1) zhuge_2003@hotmail.com
  Wei Lu (2) luwei3@mail.sysu.edu.cn
  Wei Sun (1) sunwei@mail.sysu.edu.cn
• 关键词：Image splicing detection – ; Digital image forensics – ; Approximate run length – ; Edge detection – ; Characteristic function
• 刊名：Lecture Notes in Computer Science
• 出版年：2012
• 出版时间：2012
• 年：2012
• 卷：7128
• 期：1
• 页码：349-360
• 全文大小：337.7 KB
• 参考文献：1. Chang, C.C., Lin, C.J.: LIBSVM – a library for support vector machines, http://www.csie.ntu.edu.tw/~cjlin/libsvm
  2. Chen, W., Shi, Y.Q., Su, W.: Image splicing detection using 2-d phase congruency and statistical moments of characteristic function. In: Imaging: Security, Steganography, and Watermarking of Multimedia Contents, p. 65050R (2007)
  3. Dong, J., Wang, W., Tan, T., Shi, Y.Q.: Run-Length and Edge Statistics Based Approach for Image Splicing Detection. In: Kim, H.J., Katzenbeisser, S., Ho, A.T.S. (eds.) IWDW 2008. LNCS, vol. 5450, pp. 76–87. Springer, Heidelberg (2009)
  4. Fu, D., Shi, Y.Q., Su, W.: Detection of Image Splicing Based on Hilbert-Huang Transform and Moments of Characteristic Functions with Wavelet Decomposition. In: Shi, Y.Q., Jeon, B. (eds.) IWDW 2006. LNCS, vol. 4283, pp. 177–187. Springer, Heidelberg (2006)
  5. Hsu, C.W., Chang, C.C., Lin, C.J.: A Practical Guide to Support Vector Classification. Department of Computer Science, National Taiwan University (April 2010)
  6. Hsu, Y.F., Chang, S.F.: Detecting image splicing using geometry invariants and camera characteristics consistency. In: IEEE ICME 2006, pp. 549–552 (2006)
  7. Hsu, Y.F., Chang, S.F.: Image splicing detection using camera response function consistency and automatic segmentation. In: IEEE ICME 2007, pp. 28–31 (2007)
  8. Mahdian, B., Saic, S.: A bibliography on blind methods for identifying image forgery. Signal Processing: Image Communication 25, 389–399 (2010)
  9. Potdar, V.M., Han, S., Chang, E.: A survey of digital image watermarking techniques. In: 3rd IEEE International Conference on Industrial Informatics (INDIN), pp. 709–716 (2005)
  10. Ng, T.T., Chang, S.F.: Blind detection of digital photomontage using higher order statistics. Tech. Rep. 201-2004-1, Columbia University (2004)
  11. Ng, T.T., Chang, S.F.: A data set of authentic and spliced image blocks. Tech. Rep. 203-2004-3, Columbia University (2004)
  12. Ng, T.T., Chang, S.F.: A model for image splicing. In: IEEE International Conference on Image Processing, pp. 1169–1172 (2004)
  13. Ng, T.T., Chang, S.F., Sun, Q.: Blind detection of photomontage using higher order statistics. In: IEEE ISCAS, pp. 688–691 (2004)
  14. Shi, Y.Q., Chen, C., Chen, W.: A natural image model approach to splicing detection. In: MM&Sec 2007, pp. 51–62. ACM, Dallas (2007)
  15. Shi, Y.Q., Chen, C., Xuan, G., Su, W.: Steganalysis Versus Splicing Detection. In: Shi, Y.Q., Kim, H.-J., Katzenbeisser, S. (eds.) IWDW 2007. LNCS, vol. 5041, pp. 158–172. Springer, Heidelberg (2008)
  16. Shi, Y.Q., Xuan, G., Zou, D., Gao, J., Yang, C., Zhang, Z., Chai, P., Chen, W., Chen, C.: Steganalysis based on moments of characteristic functions using wavelet decomposition, prediction-error image, and neural network. In: International Conference on Multimedia and Expo., Amsterdam, Netherlands, pp. 269–272 (2005)
• 作者单位：1. School of Software, Sun Yat-sen University, Guangzhou, 510006 China2. School of Information Science and Technology, Guangdong Key Laboratory of Information Security Technology, Sun Yat-sen University, Guangzhou, 510006 China
• ISSN：1611-3349

文摘

Image splicing is very common and fundamental in image tampering, which severely threatens the integrity and authenticity of images. As a result, there is a great need for the detection of image splicing. In this paper, an improved run length based scheme is proposed to detect this specific artifact. Firstly, the edge gradient matrix of an image is computed. Secondly, approximate run length is defined and calculated along the edge gradient direction. Thirdly, features are constructed from the related histograms of the approximate run length. Finally, support vector machine (SVM) is exploited to classify the authentic and spliced images using the constructed features. The experiment results demonstrate that the proposed approach can achieve a moderate accuracy with far less computational cost and much fewer features when compared with a similar method.