Double JPEG Detection in Mixed JPEG Quality Factors Using Deep Convolutional Neural Network

Park, Jinseok; Cho, Donghyeon; Ahn, Wonhyuk; Lee, Heung-Kyu

doi:10.1007/978-3-030-01228-1_39

Cited by 64 publications

(108 citation statements)

References 32 publications

Supporting

Mentioning

103

Contrasting

Order By: Relevance

“…Image Manipulation Segmentation. (Park et al 2018) train a network to find JPEG compression discrepancies between manipulated and authentic regions. (Zhou et al 2017; harness noise features to find inconsistencies within a manipulated image.…”

Section: Related Workmentioning

confidence: 99%

“…While a number of proposed solutions posed the problem as a classification task (Cozzolino et al 2018;Zhou et al 2017), where the goal is to classify whether a given image has been tampered with, there is great utility for solutions that are capable of detecting manipulated regions in a given image (Huh et al 2018;Zhou et al 2017;Park et al 2018;Salloum, Ren, and Kuo 2018). In this paper, we similarly treat this problem as a semantic segmentation task and adapt GANs (Goodfellow et al 2014) to generate samples to alleviate the lack of training data.…”

Section: Introductionmentioning

confidence: 99%

“…See Figure 1 for some examples. Most existing methods (Huh et al 2018;Zhou et al 2018;Park et al 2018;Salloum, Ren, and Kuo 2018) that utilize discriminative features like image metadata, noise models, or color artifacts due to, for example, Color Filter Array (CFA) inconsistencies, have failed to generalize well for these reasons.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Generate, Segment, and Refine: Towards Generic Manipulation Segmentation

Zhou

Chen

Han

et al. 2020

AAAI

View full text Add to dashboard Cite

Detecting manipulated images has become a significant emerging challenge. The advent of image sharing platforms and the easy availability of advanced photo editing software have resulted in a large quantities of manipulated images being shared on the internet. While the intent behind such manipulations varies widely, concerns on the spread of false news and misinformation is growing. Current state of the art methods for detecting these manipulated images suffers from the lack of training data due to the laborious labeling process. We address this problem in this paper, for which we introduce a manipulated image generation process that creates true positives using currently available datasets. Drawing from traditional work on image blending, we propose a novel generator for creating such examples. In addition, we also propose to further create examples that force the algorithm to focus on boundary artifacts during training. Strong experimental results validate our proposal.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Generate, Segment, and Refine: Towards Generic Manipulation Segmentation

Zhou

Chen

Han

et al. 2020

AAAI

View full text Add to dashboard Cite

show abstract

“…In this Letter, we propose end-to-end neural networks for double JPEG detection without histogram generation. By taking raw DCT coefficients, the proposed method finds the feature domain by itself and shows better performance than the state-of-the-art methods [2,3].…”

mentioning

confidence: 99%

“…Therefore, most existing methods [1][2][3] rely on the histogram and its related features, even after adopting a CNN framework. Although the CNNs based on DCT histograms performed effectively for double JPEG detection [2,3], they lose information left in raw DCT coefficients, when preprocessing the histogram, which leads to suppression of the network in learning fine-grain features.…”

mentioning

confidence: 99%

End‐to‐end double JPEG detection with a 3D convolutional network in the DCT domain

et al. 2020

View full text Add to dashboard Cite

Detection of double JPEG compression is essential in the field of digital image forensics. Although double JPEG compression detection methods have greatly improved with the development of convolutional neural networks (CNNs), they rely on handcrafted features such as discrete cosine transform (DCT) histograms. In this Letter, the authors propose an end-to-end trainable 3D CNN in the DCT domain for double JPEG compression detection. Moreover, they also propose a new type of module, called feature rescaling, to insert the quantisation table into the network suitably. The experiments show that the proposed method outperforms state-of-the-art methods.

show abstract