Visual Saliency and Extended Hidden Markov Model Based Approach for Image Splicing Detection

Abstract:

In order to improve the detection accuracy of spliced images, a new blind detection based on visual saliency was proposed in this paper. Firstly, create the edge conspicuous map by an improved OSF-based method, and extract fixations by visual attention model. Then locate those fixations on conspicuous edges by conspicuous edge positioning method. Accordingly, key feature fragments can be captured. Secondly, extract Extended Hidden Markov Model features, and reduce their dimension by SVM-RFE. Finally, support vector machine was exploited to classify the authentic and spliced images. The experimental results showed that, when testing on the Columbia image splicing detection dataset, the detection accuracy of the proposed method was 96.68%.

• Periodical: Applied Mechanics and Materials (Volumes 385-386)
• Pages: 1466-1469
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.385-386.1466
• Online since: August 2013
• Authors: Xiang Li, Xuan Jing Shen, Ying Da Lv, Hai Peng Chen
• Keywords: Blind Detection, Hidden Markov Model (HMM), Image Splicing, Visual Attention Model

References

[1] Wu Q, Li G H, Tu D, et al. A survey of blind digital image forensics technology for authenticity detection[J]. Acta Automat Sinica, 2008, 12: 1458-1466.

DOI: 10.3724/sp.j.1004.2008.01458

[2] Qu Z, Qiu G, Huang J. Detect digital image splicing with visual cues[C]/Information Hiding. Springer Berlin Heidelberg, 2009: 247-261.

DOI: 10.1007/978-3-642-04431-1_18

[3] Ng T T, Chang S F, Sun Q. A data set of authentic and spliced image blocks[J]. Columbia University, ADVENT Technical Report, 2004: 203-(2004).

[4] Lin Z, He J, Tang X, et al. Fast, automatic and fine-grained tampered JPEG image detection via DCT coefficient analysis[J]. Pattern Recognition, 2009, 42(11): 2492-2501.

DOI: 10.1016/j.patcog.2009.03.019

[5] Dong J, Wang W, Tan T, et al. Run-length and edge statistics based approach for image splicing detection[M]/Digital Watermarking. Springer Berlin Heidelberg, 2009: 76-87.

DOI: 10.1007/978-3-642-04438-0_7

[6] He Z, Sun W, Lu W, et al. Digital image splicing detection based on approximate run length[J]. Pattern Recognition Letters, 2011, 32(12): 1591-1597.

DOI: 10.1016/j.patrec.2011.05.013

[7] Itti L, Koch C, Niebur E. A model of saliency-based visual attention for rapid scene analysis[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 1998, 20(11): 1254-1259.

DOI: 10.1109/34.730558

[8] Shi Y Q, Chen C, Chen W. A natural image model approach to splicing detection[C]/Proceedings of the 9th workshop on Multimedia & security. ACM, 2007: 51-62.

DOI: 10.1145/1288869.1288878

[9] Guyon I, Weston J, Barnhill S, et al. Gene selection for cancer classification using support vector machines[J]. Machine learning, 2002, 46(1-3): 389-422.

[10] Chang C C, Lin C J. LIBSVM: a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology (TIST), 2011, 2(3): 27.

DOI: 10.1145/1961189.1961199