Automatic Tampering Detection in Spliced Images with Different Compression Levels
In this paper, we introduce a blind tampering detection method based on JPEG ghosts  capable of detecting tampering when it is created by splicing regions with different compression levels in an image. Given an image, a set of re-compressions of that image is generated and used to extract a feature vector to train a Support Vector Machine classifier. We used two different datasets in our experiments. The first one, extracted from Columbia Uncompressed Image Splicing Detection, was used to compare results with other works. Our method outperformed the previous ones when dealing with small tampered regions and similar qualities, offering hit rates above 97% (100% in the case of non-tampered images). With the second dataset, CASIA1 Tampered Image Detection Evaluation Dataset, our method offered a hit rate of 98.71% when discerning between the original and the spliced image, with just a 0.44% of non-tampered images wrongly classified.
KeywordsTampering JPEG Compression spliced images SVM
Unable to display preview. Download preview PDF.
- 1.Dirik, A.E., Bayram, S., Sencar, H.T., Memon, N.: New features to identify computer generated images. In: IEEE International Conference on Image Processing, ICIP 2007, September 16-October 19, vol. 4, pp. IV-433–IV-436 (2007)Google Scholar
- 2.Dirik, A.E., Memon, N.: Image tamper detection based on demosaicing artifacts. In: 2009 16th IEEE International Conference on Image Processing (ICIP), pp. 1497–1500 (November 2009)Google Scholar
- 5.Hsu, Y.-F., Chang, S.-F.: Detecting image splicing using geometry invariants and camera characteristics consistency. In: 2006 IEEE International Conference on Multimedia and Expo, pp. 549–552 (July 2006)Google Scholar
- 7.Li, C., Hong, L.: Adaptive fragile watermark for image authentication with tampering localization. In: 2nd International Conference on Anti-counterfeiting, Security and Identification, ASID 2008, pp. 22–25 (August 2008)Google Scholar
- 8.Li, W., Yu, N.: Rotation robust detection of copy-move forgery. In: 2010 17th IEEE International Conference on Image Processing (ICIP), pp. 2113–2116 (September 2010)Google Scholar
- 9.Li, Y.: A robust forensic method based on scale-invariance feature transform. In: 2011 International Conference on Multimedia Technology (ICMT), pp. 5246–5249 (July 2011)Google Scholar
- 10.Luo, W., Huang, J., Qiu, G.: Robust detection of region-duplication forgery in digital image. In: 18th International Conference on Pattern Recognition, ICPR 2006, vol. 4, pp. 746–749 (2006)Google Scholar
- 11.Valenzise, G., Tagliasacchi, M., Tubaro, S., Cancelli, G., Barni, M.: A compressive-sensing based watermarking scheme for sparse image tampering identification. In: 2009 16th IEEE International Conference on Image Processing (ICIP), pp. 1265–1268 (November 2009)Google Scholar
- 12.Ying, C., Yuping, W.: Exposing digital forgeries by detecting traces of smoothing. In: The 9th International Conference on Young Computer Scientists, ICYCS 2008, pp. 1440–1445 (November 2008)Google Scholar
- 13.Hu, Y.-P., Han, D.-Z.: An svd-based self-embedding watermarking method for image authentication. In: 10th IEEE Singapore International Conference on Communication Systems, ICCS 2006, pp. 1–5 (October 2006)Google Scholar
- 14.Zhang, P., Kong, X.: Detecting image tampering using feature fusion. In: International Conference on Availability, Reliability and Security, ARES 2009, pp. 335–340 (March 2009)Google Scholar