Abstract
Region splicing is a simple and common digital image tampering operation, where a chosen region from one image is composited into another image with the aim to modify the original image’s content. In this paper, we describe an effective method to expose region splicing by revealing inconsistencies in local noise levels, based on the fact that images of different origins may have different noise characteristics introduced by the sensors or post-processing steps. The basis of our region splicing detection method is a new blind noise estimation algorithm, which exploits a particular regular property of the kurtosis of nature images in band-pass domains and the relationship between noise characteristics and kurtosis. The estimation of noise statistics is formulated as an optimization problem with closed-form solution, and is further extended to an efficient estimation method of local noise statistics. We demonstrate the efficacy of our blind global and local noise estimation methods on natural images, and evaluate the performances and robustness of the region splicing detection method on forged images.
Similar content being viewed by others
Notes
If the regions are from the same image, the corresponding tampering operation is known as region cloning, which will not be considered in this work.
The projection directions need to be reflected in both horizontal and vertical directions, and the convolution is equivalent to projection assuming proper boundary handling.
We evaluated the noise levels of the raw images from the three image sets using our method, though there is no ground truth to compare. Images from the Van Hateren database have significantly lower noise levels (averaged noise standard deviation 0.25) due to its higher bit-depth and quality, while images from UCID and Kodak data bases have average estimated noise levels around 0.44 and 0.78, respectively.
All results are based on unoptimized MATLAB code running on a machine of 2.4 GHz and 4 GB RAM. This improves on our early results (Pan et al. 2012c), as we use separable random filters, and can be implemented as two consecutive 1D convolutions. This gives it an advantage in running time compared to the 2D random filters that have to be implemented as one 2D convolution step.
\(\beta \) can be computed from \(\alpha \) assuming the log noise has mean zero, c.f. Eq. (16).
All results are based on unoptimized MATLAB code running on a machine of 2.4 GHz and 4GB RAM.
This precludes methods that only classify whole image as containing spliced regions (e.g., Bayram et al. 2006; Fu et al. 2007; Ng and Chang 2004), or require initial user input for possible spliced regions (Popescu and Farid 2004; Hsu and Chang 2006; Lin et al. 2005, or predicate on more detailed knowledge of the imaging processes (Chen et al. 2007).
References
Andrews, D. F., & Mallows, C. L. (1974). Scale mixtures of normal distributions. Journal of the Royal Statistical Society, Series B, 36(1), 99–102.
Bayram, S., Avcibas, I., Sankur, B., & Memon, N. D. (2006). Image manipulation detection. Journal of Electronic Imaging, 15(4), 1–17.
Benedict, T. R., & Soong, T. T. (1967). The joint estimation of signal and noise from the sum envelope. IEEE Transactions on Information Theory, 13(3), 447–454.
Bethge, M. (2006). Factorial coding of natural images: How effective are linear models in removing higher-order dependencies? The Journal of the Optical Society of America A, 23(6), 1253–1268.
Bilcu, R. C., & Vehvilainen, M. (2005). A new method for noise estimation in images. In IEEE EURASIP International workshop on nonlinear signal and image processing.
Burt, P., & Adelson, E. (1981). The Laplacian pyramid as a compact image code. IEEE Transactions on Communication, 31(4), 532–540.
Cardoso, J. F. (1999). High-order contrasts for independent component analysis. Neural Computation, 11(1), 157–192.
Chen, M., Fridrich, J. J., Lukás, J., & Goljan, M. (2007). Imaging sensor noise as digital X-ray for revealing forgeries. In Information hiding (pp. 342–358).
Chen, W., Shi, Y. Q., & Su, W. (2007). Image splicing detection using 2-D phase congruency and statistical moments of characteristic function. In Society of photo-optical instrumentation engineers (SPIE) conference series (Vol. 6505). doi:10.1117/12.704321.
Chen, M., Fridrich, J., Goljan, M., & Lukas, J. (2008). Determining image origin and integrity using sensor noise. IEEE Transactions on Information Forensics and Security, 3(1), 74–90. doi:10.1109/TIFS.2007.916285.
Crow, F. (1984). Summed-area tables for texture mapping. In ACM SIGGRAPH.
Danielyan, A., & Foi, A. (2009). Noise variance estimation in nonlocal transform domain. In International workshop on local and non-local approximation in image processing.
Donoho, D. L. (1995). De-noising by soft-thresholding. IEEE Transactions on Information Theory, 41(3), 613–627.
Farebrother, R. W. (1990). The cumulants of the logarithm of a gamma variable the cumulants of the logarithm of a gamma variable. Journal of Statistical Computation and Simulation, 36(4), 243–245.
Farid, H. (2009). Photo fakery and forensics. Advances in Computers, 77, 1–55.
Feller, W. (1968). An introduction to probability theory and its applications. New York: Wiley.
Field, D. J. (1987). Relations between the statistics of natural images and the response properties of cortical cells. The Journal of the Optical Society of America A, 4(12), 2379–2394.
Filler, T., Fridrich, J. J., & Goljan, M. (2008). Using sensor pattern noise for camera model identification. In IEEE International conference on image processing. San Diego, CA.
Foi, A., Alenius, S., Katkovnik, V., & Egiazarian, K. (2007). Noise measurement for raw-data of digital imaging sensors by automatic segmentation of non-uniform targets. IEEE Sensors Journal, 7(10), 1456–1461.
For̈stner, W. (1998). Image preprocessing for feature extraction in digital intensity, color and range images. In: Springer lecture notes on Earth siences.
Franzen, R. (1999). Kodak lossless true color image suite. http://r0k.us/graphics/kodak. Accessed 17 Dec 2013.
Fu, D., Shi, Y. Q., & Su, W. (2007). Image splicing detection using 2-d phase congruency and statistical moments of characteristic function. In Proceedings of SPIE security, steganography, and watermarking of multimedia contents IX.
Garry, M., & Gerrie, M. P. (2005). When photographs create false memories. Current Directions in Psychological Science, 14, 326–330.
He, J., Lin, Z., Wang, L., & Tang., X. (2006). Detecting doctored JPEG images via DCT coefficient analysis. In ECCV.
Hsu, Y. F., & Chang, S. F. (2006). Detecting image splicing using geometry invariants and camera characteristics consistency. In IEEE International Conference on Multimedia and Expo.
Hsu, Y. F., & Chang, S. F. (2007). Image splicing detection using camera response function consistency and automatic segmentation. In IEEE International Conference on Multimedia and Expo.
Hyvärinen, A. (1999). Fast and robust fixed-point algorithms for independent component analysis. IEEE Transactions on Neural Networks, 10(3), 626–634.
Konstantinides, G. S. Y. K., & Natarajan, B. K. (1997). Noise estimation and filtering using block-based singular value decomposition. IEEE Transactions on Image Processing, 6(3), 479–483.
Lin, Z., Wang, R., Tang, X., & Shum, H. (2005). Detecting doctored images using camera response normality and consistency. In CVPR.
Lin, Z., He, J., Tang, X., & Tang, C. (2009). Fast, automatic and fine-grained tampered JPEG images detection via DCT coefficient analysis. Pattern Recognition, 42(11), 2492–2501.
Liu, C., Szeliski, R., Kang, S. B., Zitnick, C. L., & Freeman, W. T. (2008). Automatic estimation and removal of noise from a single image. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(2), 299–314. http://doi.ieeecomputersociety.org/10.1109/TPAMI.2007.1176.
Liu, X., Tanaka, M., Okutomi, M. (2012). Noise level estimation using weak textured patches of a single noisy image. In IEEE International conference on image processing.
Lukas, J., Fridrich, J., & Goljan, M. (2006). Detecting digital image forgeries using sensor pattern noise. In Proceedings of SPIE security, steganography, and watermarking of multimedia contents VIII.
Lukás, J., Fridrich, J. J., & Goljan, M. (2006). Digital camera identification from sensor pattern noise. IEEE Transactions on Information Forensics and Security, 1(2), 205–214.
Lyu, S., & Simoncelli, E. P. (2009). Nonlinear extraction of ’independent components’ of natural images using radial Gaussianization. Neural Computation, 18(6), 1–35.
Mahdian, B., & Saic, S. (2009). Using noise inconsistencies for blind image forensics. Image and Vision Computing, 27(10), 1497–1503.
Matzner, R., & Engleberger, F. (1994). An SNR estimation algorithm using fourth-order moments. In IEEE International symposium on information theory.
Meer, P., Jolion, J., & Rosenfeld, A. (1990). A fast parallel algorithm for blind estimation of noise variance. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(2), 216–223.
Nakamura, J. (Ed.). (2006). Image sensors and signal processing for digital still cameras. Boca Raton: Taylor and Francis.
Ng, T. T., & Chang, S. F. (2004). A model for image splicing. In IEEE International conference on image processing (ICIP). Singapore.
Olsen, S. I. (1993). Estimation of noise in images: An evaluation. Computer Vision Graphics Image Process. Graphic Models and Image Process, 55(4), 319–323.
Pan, X., Zhang, X., & Lyu, S. (2011). Exposing image forgery with blind noise estimation. In The 13th ACM workshop on multimedia and security. Buffalo, NY.
Pan, X., Zhang, X., & Lyu, S. (2012). Blind local noise estimation for medical images reconstructed from rapid acquisition. In SPIE Symposium on medical imaging. San Diego, CA.
Pan, X., Zhang, X., & Lyu, S. (2012). Detecting splicing in digital audios using local noise level estimation. In IEEE International conference on acoustics, speech, and signal processing (ICASSP). Kyoto, Japan.
Pan, X., Zhang, X., & Lyu, S. (2012). Exposing image splicing with inconsistent local noise variances. In IEEE International conference on computational photography. Seattle, WA.
Pauluzzi, D. R., & Beaulieu, N. C. (2000). A comparison of SNR estimation techniques for the AGWN channel. IEEE Transactions on Communications, 48(10), 1681–1691.
Ponomarenko, N. N., Lukin, V. V., Abramov, S. K., Egiazarian, K. O., & Astola, J. T. (2003). Blind evaluation of additive noise variance in textured images by nonlinear processing of block DCT coefficients. In SPIE Proceedings (Vol. 5014, pp. 178–189).
Popescu, A., & Farid, H. (2004). Statistical tools for digital forensics. In 6th International workshop on information hiding. Toronto, Canada.
Portilla, J. (2004). Full blind denoising through noise covariance estimation using Gaussian scale mixtures in the wavelet domain. In International conference on image processing. doi:10.1109/ICIP.2004.1419524.
Portilla, J., Strela, V., Wainwright, M. J., & Simoncelli, E. P. (2003). Image denoising using scale mixtures of Gaussians in the wavelet domain. IEEE Transactions on Image Processing, 12(11), 1338–1351.
Qu, Z., Qiu, G., & Huang, J. (2009). Detect digital image splicing with visual cues. In: S. Katzenbeisser, & A. R. Sadeghi (Eds.) International workshop on information hiding (pp. 247–261).
Rank, K., Lendl, M., & Unbehauen, R. (1999). Estimation of image noise variance. In IEE Proceedings. Vision, Image and Signal Processing (Vol. 146, pp. 80–84).
Rudin, L., Lions, P., & Osher, S. (2003). Multiplicative denoising and deblurring: Theory and algorithms. In S. Osher & N. Paragios (Eds.), Geometric level set methods in imaging, vision, and graphics. New York: Springer.
Sacchi, D. L. M., Agnoli, F., & Loftus, E. F. (2007). Changing history: Doctored photographs affect memory for past public events. Applied Cognitive Psychology, 21, 1005–1022.
Schaefer, G., & Stich, M. (2004). UCID—an uncompressed colour image database. In Proc. SPIE, storage and retrieval methods and applications for multimedia.
Schmidt, U., Schelten, K., & Roth, S. (2011). Bayesian deblurring with integrated noise estimation. In IEEE International conference on computer vision. Colorado Springs, CO.
Sencar, H. T., & Memon, N. (Eds.). (2012). Digital image forensics: There is more to a picture than meets the eye. Dordrecht: Springer.
Serra, J. (1988). Image analysis and mathematical morphology: Theoretical advances. Image Analysis and Mathematical Morphology. London: Academic Press.
Shi, Y. Q., Chen, C., & Chen, W. (2007). A natural image model approach to splicing detection. In Proceedings of the 9th workshop on multimedia and security (pp. 51–62). New York, NY: ACM. doi:10.1145/1288869.1288878.
Simoncelli, E. P., & Freeman, W. T. (1995). The steerable pyramid: A flexible architecture for multi-scale derivative computation. In IEEE Second international conference on image processing.
Simoncelli, E. P., & Olshausen, B. (2001). Natural image statistics and neural representation. Annual Review of Neuroscience, 24, 1193–1216. doi:10.1146/annurev.neuro.24.1.1193.
Stefano, A., White, P., & Collis, W. (2004). Training methods for image noise level estimation on wavelet components. EURASIP Journal on Applied Signal Processing, 16, 2400–2407.
Tai, S. C., & Yang, S. M. (2008). A fast method for image noise estimation using Laplacian operator and adaptive edge detection. In International symposium on communications, control and signal processing.
van Hateren, J. H., & van der Schaaf, A. (1998). Independent component filters of natural images compared with simple cells in primary visual cortex. Proceedings of the Royal Society B: Biological Sciences, 265(1394), 359–366.
Viola, P., & Jones, M. (2002). Robust real-time object detection. International Journal of Computer Vision, 57(2), 137–154.
Wainwright, M. J., & Simoncelli, E. P. (2000). Scale mixtures of Gaussians and the statistics of natural images. Cambridge, MA: MIT Press.
Wang, W., Dong, J., & Tan, T. (2009). Effective image splicing detection based on image chroma. In IEEE International conference on image processing.
Withagen, P., Groen, F., & Schutte, K. (2005). CCD characterization for a range of color cameras. In Instrumentation and measurement technology conference, 2005 (Vol 3, pp. 2232–2235). doi:10.1109/IMTC.2005.1604573.
Zoran, D., & Weiss, Y. (2009). Scale invariance and noise in nature image. In IEEE International conference on computer vision. Kyoto, Japan.
Acknowledgments
We would like to thank Daniel Zoran, Zhouchen Lin and Babak Mahdian for kindly sharing the images, codes and results of their works with us. We would also like to thank the two anonymous reviewers for their constructive comments that helped us improve this work. This work is supported in part by the National Science Foundation under Grant Nos. IIS-0953373, IIS-1208463 and CCF-1319800.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Lyapunov Central Limit Theorem
Suppose \(\{X_1, \cdots , X_n\}\) is a sequence of independent random variables, each with finite expected value \(\mu _i\) and variance \(\sigma ^2_i\). Define \(s_n^2 = \sum _{i=1}^n \sigma _i^2\), If for some \(\delta > 0\), the ”Lyapunovs condition”
is satisfied, then a sum of \((X_i - \mu _i)/s_n\) converges in distribution to a standard normal random variable, as \(n\) goes to infinity:
Appendix 2: Derivation of Claim 1
Based on the joint density function of the GSM variable \(\mathbf {x}\), Eq. (5), we can obtain the marginal distribution of its projection on a non-zero vector \(\mathbf {w}\) as:
The marginalization in the inside integral reduces to a Gaussian distribution with zero mean and variance \(\mathrm {z}\mathbf {w}^{\scriptscriptstyle T}\Sigma _\mathbf {x}\mathbf {w}\), based on the property of Gaussian distributions. Therefore, \(p_\mathbf {w}(t)\) is a 1D GSM model with mixing density \(p_\mathrm {z}(\mathrm {z})\), as:
Now, the variance of \(\mathbf {w}^{\scriptscriptstyle T}\mathbf {x}\) is computed as
Furthermore, the fourth order moment of \(\mathbf {w}^{\scriptscriptstyle T}\mathbf {x}\) is given as
where we use the fact that for a Gaussian distribution \(\mathcal{N}_{t}(0,\sigma ^2)\), its fourth order moment is \(3\sigma ^4\). Putting all results together in the definition of kurtosis, we have
Appendix 3: Derivation of Eq. (11)
First we expand the objective function in (7) as
The gradient of \(L\left( \sqrt{\kappa },\sigma ^2\right) \) with regards to the two parameters are computed as, as:
Setting Eq. (18) to zero, and considering \(\sqrt{\kappa } > 0\), we have
Setting Eq. (19) to zero and substituting with Eq. (20) yield
from which we can obtain
Next, substituting Eq. (21) back into Eq. (20), we have
Further arranging terms and replacing average over different channels with \(\langle \cdot \rangle _k\) yield Eq. (11). Further checking the second-order conditions ensures that the solution is the unique global minimizer of Eq. (7).
Rights and permissions
About this article
Cite this article
Lyu, S., Pan, X. & Zhang, X. Exposing Region Splicing Forgeries with Blind Local Noise Estimation. Int J Comput Vis 110, 202–221 (2014). https://doi.org/10.1007/s11263-013-0688-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11263-013-0688-y