Abstract
Photo-response non-uniformity (PRNU), as a class of device fingerprint, plays a key role in the forgery detection/localization for visual media. The state-of-the-art PRNU-based forensics methods generally rely on the multi-scale trace analysis and result fusion, with Markov random field model. However, such hand-crafted strategies are difficult to provide satisfactory multi-scale decision, exhibiting a high false-positive rate. Motivated by this, we propose an end-to-end multi-scale decision fusion strategy, where a mapping from multi-scale forgery probabilities to binary decision is achieved by a supervised deep fully connected neural network. As the first time, the deep learning technology is employed in PRNU-based forensics for more flexible and reliable integration of multi-scale information. The benchmark experiments exhibit the state-of-the-art accuracy performance of our method in both pixel-level and image-level, especially for false positives. Additional robustness experiments also demonstrate the benefits of the proposed method in resisting noise and compression attacks.
- [1] . 2016. Media forensics. In Information Hiding. Artech House, 231–259.Google Scholar
- [2] . 2017. Splicing image forgery detection using textural features based on the grey level co-occurrence matrices. IET Image Process. 11, 1 (2017), 44–53.Google Scholar
Cross Ref
- [3] . 2009. Analysis of denoising filters for photo response non uniformity noise extraction in source camera identification. In Proceedings of the IEEE International Conference on Digital Signal Processing. 1–7.Google Scholar
Cross Ref
- [4] . 2011. A sift-based forensic method for copy–move attack detection and transformation recovery. IEEE Trans. Inf. Forens. Secur. 6, 3 (2011), 1099–1110.Google Scholar
Digital Library
- [5] . 2016. Behavior knowledge space-based fusion for copy–move forgery detection. IEEE Trans. Image Process. 25, 10 (2016), 4729–4742.Google Scholar
Digital Library
- [6] . 2015. Efficient dense-field copy–move forgery detection. IEEE Trans. Inf. Forens. Secur. 10, 11 (2015), 2284–2297.Google Scholar
Digital Library
- [7] . 2012. An evaluation of popular copy-move forgery detection approaches. IEEE Trans. Inf. Forens. Secur. 7, 6 (2012), 1841–1854.Google Scholar
Digital Library
- [8] . 2019. Noiseprint: A CNN-based camera model fingerprint. IEEE Trans. Inf. Forens. Secur. 15 (2019), 144–159.Google Scholar
Cross Ref
- [9] . 2020. On addressing the impact of ISO speed upon PRNU and forgery detection. IEEE Trans. Inf. Forens. Secur. 16 (2020), 190–202.Google Scholar
Digital Library
- [10] . 2020. Camera fingerprint extraction via spatial domain averaged frames. IEEE Trans. Inf. Forens. Secur. 15 (2020), 3270–3282.Google Scholar
Cross Ref
- [11] . 2021. A serial image copy-move forgery localization scheme with source/target distinguishment. IEEE Trans. Multimedia 23 (2021), 3506–3517.
DOI: Google ScholarDigital Library
- [12] . 2021. Image tampering localization using a dense fully convolutional network. IEEE Trans. Inf. Forens. Secur. 16 (2021), 2986–2999.Google Scholar
Cross Ref
- [13] . 2020. Controlling neural learning network with multiple scales for image splicing forgery detection. ACM Trans. Multimed. Comput. Commun. Appl. 16, 4 (2020), 1–22.Google Scholar
Digital Library
- [14] . 2018. Contrast enhancement estimation for digital image forensics. ACM Trans. Multimed. Comput. Commun. Appl. 14, 2 (2018), 1–21.Google Scholar
Digital Library
- [15] . 2017. Secure cloud-based image tampering detection and localization using POB number system. ACM Trans. Multimed. Comput. Commun. Appl. 13, 3 (2017), 1–23.Google Scholar
Digital Library
- [16] . 2013. A generalized tamper localization approach for reversible watermarking algorithms. ACM Trans. Multimed. Comput. Commun. Appl. 9, 3 (2013), 1–22.Google Scholar
Digital Library
- [17] . 1994. Radiometric CCD camera calibration and noise estimation. IEEE Trans. Pattern Anal. Mach. Intell. 16, 3 (1994), 267–276.Google Scholar
Digital Library
- [18] . 2013. Sensor defects in digital image forensic. In Digital Image Forensics. Springer, 179–218.Google Scholar
Cross Ref
- [19] . 2006. Digital camera identification from sensor pattern noise. IEEE Trans. Inf. Forens. Secur. 1, 2 (2006), 205–214.Google Scholar
Digital Library
- [20] . 2013. Improved photo response non-uniformity (PRNU) based source camera identification. Forens. Sci. Int. 226, 1-3 (2013), 132–141.Google Scholar
Cross Ref
- [21] . 2014. A context-adaptive SPN predictor for trustworthy source camera identification. EURASIP J. Image Video Process. 2014, 1 (2014), 1–11.Google Scholar
Cross Ref
- [22] . 2016. Fast source camera identification using content adaptive guided image filter. J. Forens. Sci. 61, 2 (2016), 520–526.Google Scholar
Cross Ref
- [23] . 2003. Higher-order wavelet statistics and their application to digital forensics. In Proceedings of the IEEE Computer Society Conference on Computer Vision Pattern Recognition Workshops, Vol. 8. 94–94.Google Scholar
Cross Ref
- [24] . 2017. Forensics feature analysis in quaternion wavelet domain for distinguishing photographic images and computer graphics. Multimed. Tools Appl. 76, 22 (2017), 23721–23737.Google Scholar
Digital Library
- [25] . 2018. Analytical global median filtering forensics based on moment histograms. ACM Trans. Multimed. Comput. Commun. Appl. 14, 2 (2018), 1–23.Google Scholar
Digital Library
- [26] . 2014. A feature-based approach for image tampering detection and localization. In Proceedings of the IEEE International Workshop on Information Forensics Security. 149–154.Google Scholar
Cross Ref
- [27] . 2016. Identification of various image operations using residual-based features. IEEE Trans. Circ. Syst. Video Technol. 28, 1 (2016), 31–45.Google Scholar
Digital Library
- [28] . 2014. Weighted averaging-based sensor pattern noise estimation for source camera identification. In Proceedings of the IEEE International Conference on Image Processing. 5357–5361.Google Scholar
Cross Ref
- [29] . 2011. Enhancing source camera identification performance with a camera reference phase sensor pattern noise. IEEE Trans. Inf. Forens. Secur. 7, 2 (2011), 393–402.Google Scholar
Digital Library
- [30] . 2015. Preprocessing reference sensor pattern noise via spectrum equalization. IEEE Trans. Inf. Forens. Secur. 11, 1 (2015), 126–140.Google Scholar
Digital Library
- [31] . 2006. Detecting digital image forgeries using sensor pattern noise. In Security, Steganography, and Watermarking of Multimedia Contents VIII, Vol. 6072, Edward J. Delp III and Ping Wah Wong (Eds.). SPIE, 362–372.
DOI: Google ScholarCross Ref
- [32] . 2008. Determining image origin and integrity using sensor noise. IEEE Trans. Inf. Forens. Secur. 3, 1 (2008), 74–90.Google Scholar
Digital Library
- [33] . 2011. PRNU-based detection of small-size image forgeries. In Proceedings of the IEEE International Conference on Digital Signal Processing. 1–6.Google Scholar
Cross Ref
- [34] . 2014. Guided filtering for PRNU-based localization of small-size image forgeries. In Proceedings of the IEEE International Conference on Acoustic Speech Signal Processing. 6231–6235.Google Scholar
Cross Ref
- [35] . 2012. Guided image filtering. IEEE Trans. Pattern Anal. Mach. Intell. 35, 6 (2012), 1397–1409.Google Scholar
Digital Library
- [36] . 2016. Multi-scale analysis strategies in PRNU-based tampering localization. IEEE Trans. Inf. Forens. Secur. 12, 4 (2016), 809–824.Google Scholar
Digital Library
- [37] . 2019. Multi-scale segmentation strategies in PRNU-based image tampering localization. Multimed. Tools Appl. 78, 14 (2019), 20113–20132.Google Scholar
Digital Library
- [38] . 2007. Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans. Image Process. 16, 8 (2007), 2080–2095.Google Scholar
Cross Ref
- [39] . 2016. DNN speaker adaptation using parameterised sigmoid and ReLU hidden activation functions. In Proceedings of the IEEE International Conference on Acoustic Speech Signal Processing. 5300–5304.Google Scholar
Digital Library
- [40] . 2017. Data classification with deep learning using Tensorflow. In Proceedings of the IEEE International Conference on Computer Science and Engineering (UBMK). 755–758.Google Scholar
Cross Ref
- [41] . 2018. Comparison of non-linear activation functions for deep neural networks on MNIST classification task. ArXiv Preprint ArXiv:1804.02763 (2018).Google Scholar
- [42] . 2019. Dynamic weighted cross entropy for semantic segmentation with extremely imbalanced data. In Proceedings of the IEEE International Conference on Artificial Intelligence and Advanced Manufacturing. 230–233.
DOI: Google ScholarCross Ref
- [43] . 2019. FU-net: Multi-class image segmentation using feedback weighted U-net. In Lecture Notes in Computer Science. Springer, 529–537.Google Scholar
- [44] . 2017. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In Lecture Notes in Computer Science.Springer, 240–248.Google Scholar
- [45] . 2015. Raise: A raw images dataset for digital image forensics. In Proceedings of the ACM Multimedia Systems Conference.219–224.Google Scholar
Digital Library
Index Terms
PRNU-based Image Forgery Localization with Deep Multi-scale Fusion
Recommendations
JPEG Compression-aware Image Forgery Localization
MM '22: Proceedings of the 30th ACM International Conference on MultimediaImage forgery localization, which aims to find suspicious regions tampered with splicing, copy-move or removal manipulations, has attracted increasing attention. Existing image forgery localization methods have made great progress on public datasets. ...
Image Forgery Localization based on Multi-Scale Convolutional Neural Networks
IH&MMSec '18: Proceedings of the 6th ACM Workshop on Information Hiding and Multimedia SecurityIn this paper, we propose to utilize Convolutional Neural Networks (CNNs) and the segmentation-based multi-scale analysis to locate tampered areas in digital images. First, to deal with color input sliding windows of different scales, we adopt a unified ...
An efficient method for image forgery detection based on trigonometric transforms and deep learning
AbstractImage forgery detection is the basic key to solve many problems, especially social problems such as those in Facebook, and court cases. The common form of image forgery is the copy-move forgery, in which a section of the image is copied and pasted ...






Comments