Deep Learning for Detecting Processing History of Images

Boroumand, Mehdi; Fridrich, Jessica

doi:10.2352/issn.2470-1173.2018.07.mwsf-213

Cited by 50 publications

(37 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We manipulated 50% of the image patch pairs with the same editing operation, but with different parameter, and manipulated 50% of the pairs with different editing operations. The known manipulations were the same manipulations used in [6] and [8]. We repeated this for the evaluation database, using both the "known" and "unknown" manipulations.…”

Section: B Editing Operation Comparisonmentioning

confidence: 99%

“…Recently, researchers have developed deep learning methods that target digital image forensic tasks with high accuracy. For example, convolutional neural network (CNN) based systems have been proposed that accurately detect traces of median filtering [2], resizing [3], [4], inpainting [5], multiple processing operations [6]- [8], processing order [9], and double JPEG compression [10], [11]. Additionally, researchers have proposed approaches to identify the source camera model of digital images [12]- [15], and identify their origin social media website [16].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Forensic Similarity for Digital Images

Mayer

Stamm

2020

IEEE Trans.Inform.Forensic Secur.

114

View full text Add to dashboard Cite

In this paper we introduce a new digital image forensics approach called forensic similarity, which determines whether two image patches contain the same forensic trace or different forensic traces. One benefit of this approach is that prior knowledge, e.g. training samples, of a forensic trace are not required to make a forensic similarity decision on it in the future. To do this, we propose a two part deeplearning system composed of a CNN-based feature extractor and a three-layer neural network, called the similarity network. This system maps pairs of image patches to a score indicating whether they contain the same or different forensic traces. We evaluated system accuracy of determining whether two image patches were 1) captured by the same or different camera model, 2) manipulated by the same or different editing operation, and 3) manipulated by the same or different manipulation parameter, given a particular editing operation. Experiments demonstrate applicability to a variety of forensic traces, and importantly show efficacy on "unknown" forensic traces that were not used to train the system. Experiments also show that the proposed system significantly improves upon prior art, reducing error rates by more than half. Furthermore, we demonstrated the utility of the forensic similarity approach in two practical applications: forgery detection and localization, and database consistency verification.

show abstract

Section: B Editing Operation Comparisonmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Forensic Similarity for Digital Images

Mayer

Stamm

2020

IEEE Trans.Inform.Forensic Secur.

114

View full text Add to dashboard Cite

show abstract

“…Experiments with fixed, constrained, and randomly initialised kernels led Boroumand and Fridrich [ 28 ] to the notion that no constraints of any kind should be imposed on the filters from the first layer. According to them the fixed or constrained kernels remove information about the image luminance, which can be damaging for example when trying to detect luminance adjustments, such as gamma corrections and brightness and contrast changes.…”

Section: Cnn Architecturementioning

confidence: 99%

Interpol review of imaging and video 2016–2019

Geradts

Filius

Ruifrok

2020

Forensic Science International: Synergy

View full text Add to dashboard Cite

show abstract

“…In [123], a CNN was trained to identify laundering techniques applied to an image. The laundering techniques, applied singly, were: low-pass-, high-…”

Section: A N U S C R I P Tmentioning

confidence: 99%

A review of digital video tampering: From simple editing to full synthesis

Johnston

Elyan

2019

Digital Investigation

View full text Add to dashboard Cite

Video tampering methods have witnessed considerable progress in recent years. This is partly due to the rapid development of advanced deep learning methods, and also due to the large volume of video footage that is now in the public domain. Historically, convincing video tampering has been too labour intensive to achieve on a large scale. However, recent developments in deep learning-based methods have made it possible not only to produce convincing forged video but also to fully synthesize video content. Such advancements provide new means to improve visual content itself, but at the same time, they raise new challenges for state-of-the-art tampering detection methods. Video tampering detection has been an active field of research for some time, with periodic reviews of the subject. However, little attention has been paid to video tampering techniques themselves. This paper provides an objective and in-depth examination of current techniques related to digital video manipulation. We thoroughly examine their development, and show how current evaluation techniques provide opportunities for the advancement of video tampering detection. A critical and extensive review of photo-realistic video synthesis is provided with emphasis on deep learning-based methods. Existing tampered video datasets are also qualitatively reviewed and critically discussed. Finally, conclusions are drawn upon an exhaustive and thorough review of tampering methods with

show abstract

Deep Learning for Detecting Processing History of Images

Cited by 50 publications

References 33 publications

Forensic Similarity for Digital Images

Forensic Similarity for Digital Images

Interpol review of imaging and video 2016–2019

A review of digital video tampering: From simple editing to full synthesis

Contact Info

Product

Resources

About