Abstract
Seam carving is a content-aware algorithm for image resizing and tampering. This algorithm assigns an energy map to an image and removes the seams with low energy from the image. By doing this, seam carving makes it possible to reduce the image size and eliminate specific content from images. The detection of seam carving has lately been an important but challenging area of research. In past work, we had proposed a method that involved analyzing optimal patch types to recover seams and thus to detect seam-carved images. This method yielded highly accurate detection results. In this paper, we introduce an auto-associated Hopfield network to determine the optimal patch type for seam recovery. We use the Hebbian learning rule to choose, among candidate patch types, the one that most closely resembles the relevant target pattern. Experiments showed that the retrieval process usually converged within eight iterations and that the converged patterns improved the detection accuracy, e.g., with rates of 95.97 and 98.55% for 20 and 50% seam-carved images respectively. We also used this enhanced patch analysis method to identify the seam-carved regions of a tampered image. Its accuracy for the identification of tampered regions was higher than 70% for images with < 30% seam carving.
Similar content being viewed by others
References
Dong W, Zhou N, Paul JC, Zhang X (2009) Optimized image resizing using seam carving and scaling. ACM Trans Graph 28(10):125:1–125:10
Elliott T (2003) An analysis of synaptic normalization in a general class of hebbian models. Neural Comput 15(4):937–963
Fillion C, Sharma G (2010) Detecting content adaptive scaling of images for forensic applications. Proc SPIE: Media Forensics Secur 7541:36–47
Galtier MN, Faugeras OD, Bressloff PC (2012) Hebbian learning of recurrent connections: a geometrical perspective. Neural Comput 24(9):2346–2383
Gonzalez RC, Woods RE (2001) Digital image processing, 2nd edn. Prentice Hall, Englewood Cliffs
Goodall TR, Katsavounidis I, Li Z, Aaron A, Bovik AC (2016) Blind picture upscaling ratio prediction. IEEE Signal Process Lett 23:1801–1805
Jacyna GM, Malaret ER (1989) Classification performance of a hopfield neural network based on a hebbian-like learning rule. IEEE Trans Inf Theory 35(2):263–280
Krizhevsky A, Sutskever I, Hinton GE (2017) Imagenet classification with deep convolutional neural networks. Commun ACM 60(6):84–90
Li ZN, Drew MS, Liu J (2014) Image compression standards. In: Fundamentals of multimedia, 2 edn, chap. 9. Springer, Berlin, pp 281–315
Liu Q (2016) An approach to detecting jpeg down-recompression and seam carving forgery under recompression anti-forensics. Pattern Recognit 65:35–46
Liu Q, Chen Z (2014) Improved approaches with calibrated neighboring joint density to steganalysis and seam-carved forgery detection in jpeg images. ACM Trans Intell Syst Technol 5(4):63:1–63:30
Liu Q, Sung AH, Qiao M (2011) Neighboring joint density-based jpeg steganalysis. ACM Trans Intell Syst Technol 2(2):16:1–16:16
Lu W, Varna AL, Wu M (2010) Forensic hash for multimedia information. In: SPIE Media Forensics and Security, pp 75410–75419
Palmieri F, Zhu J (1995) Self-association and hebbian learning in linear neural networks. IEEE Trans Neural Netw 6(5):1165–1184
Rubinstein M, Gutierrez D, Sorkine O, Shamir A (2010) A comparative study of image retargeting. ACM Trans Graph 29:160:1–160:10
Rubinstein M, Shamir A, Avidan S (2008) Improved seam carving for video retargeting. ACM Trans Graph 27(3):16:1–16:9
Ryu SJ, Lee HY, Lee HK (2013) Detection of content-aware image resizing using seam properties. Appl Mech Mater 284:3074–3078
Sarkar A, Nataraj L, Manjunath BS (2009) Detection of seam carving and localization of seam insertions in digital images. In: Proceedings of 11th ACM workshop on multimedia and security, pp 107–116
Schaefer G, Stich M (2003) UCID: an uncompressed color image database. In: Proceedings of SPIE 5307, pp 472–480
Shamir A, Avidan S (2007) Seam carving for content aware image resizing. ACM Trans Graph 26(3):107–216
Sheng G, Li T, Su Q, Chen B, Tang Y (2016) Detection of content-aware image resizing based on benfords law. Soft Computing, pp 1–9
Shi Y.Q, Chen C, Chen W (2006) A markov process based approach to effective attacking jpeg steganography. In: Lecture notes in computer science, pp 249–264
Wang Y, Liu J, Li Y, Yan J, Lu H (2016) Objectness-aware semantic segmentation. In: Proceedings of the 2016 ACM on multimedia conference (ACM MM 2016), pp 307–311
Wattanachote K, Shih TK, Chang WL, Chang HH (2015) Tamper detection of jpeg image due to seam modifications. IEEE Trans Inf Forensics Secur 10(12):2477–2491
Wei JD, Lin YJ, Wu YJ (2014) A patch analysis method to detect seam carved images. Pattern Recognit Lett 36:100–106
Wei JD, Lin YJ, Wu YJ, Kang LW (2013) A patch analysis approach for seam carved image detection. In: Proceedings of 40th international conference and exhibition on computer graphics and interactive techniques (ACM SIGGRAPH 2013)
Yan B, Yang X, Li K (2014) Efficient image retargeting via adaptive pixel fusion. In: Proceedings of ACM international conference on multimedia, pp 929–932
Ye J, Shi YQ (2017) An effective method to detect seam carving. J Inf Secur Appl 35:13–22
Yin T, Yang G, Li L, Zhang D, Sun X (2015) Detecting seam carving based image resizing using local binary patterns. Comput Secur 55:130–141
Zhang D, Li Q, Yang G, Li L, Sun X (2017) Detection of image seam carving by using weber local descriptor and local binary patterns. J Inf Secur Appl 36:135–144
Zhang D, Yin T, Yang G, Xia M, Li L, Sun X (2017) Detecting image seam carving with low scaling ratio using multi-scale spatial and spectral entropies. J Vis Commun Image Represent 48:281–291
Zhu N, Deng C, Gao X (2016) A learning-to-rank approach for image scaling factor estimation. Neurocomputing 204(C):33–40
Acknowledgements
This work was supported in part by the Ministry of Science and Technology, Taiwan, R.O.C. (Grant Nos. MOST 103-2221-E-182-049 and MOST-106-2221-E-182-075) and Chang Gung Memorial Hospital (Grant No. BMRPB21).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no potential conflicts of interest with respect to the research, authorship, and publication of this article.
Rights and permissions
About this article
Cite this article
Wei, JD., Cheng, HJ. & Chang, CW. Hopfield network-based approach to detect seam-carved images and identify tampered regions. Neural Comput & Applic 31, 6479–6492 (2019). https://doi.org/10.1007/s00521-018-3463-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-018-3463-8