Image Manipulation Detection and Localization Based on the Dual-Domain Convolutional Neural Networks

Shi, Zenan; Shen, Xuanjing; Kang, Hui; Lv, Yingda

doi:10.1109/access.2018.2883588

Cited by 36 publications

(18 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We finally compare the localization performance of our method with several state-of-the-art deep learning based methods in Shi [40], Huh [41] and Salloum [43]. In the interest of fairness, following [40] and [43], we train our method on CASIA v2.0 dataset while evaluate the localization performance on DSO-1 dataset. We use the results reported in [40], [41] and [43] for comparison.…”

Section: ) Comparison With Other State-of-the-art Splicing Localizatmentioning

confidence: 99%

“…In the interest of fairness, following [40] and [43], we train our method on CASIA v2.0 dataset while evaluate the localization performance on DSO-1 dataset. We use the results reported in [40], [41] and [43] for comparison. As illustrated in Table 7, the proposed method outperforms Huh [41] and Salloum [43] and achieves comparable performance to [40] in terms of F 1 -score.…”

Section: ) Comparison With Other State-of-the-art Splicing Localizatmentioning

confidence: 99%

“…We use the results reported in [40], [41] and [43] for comparison. As illustrated in Table 7, the proposed method outperforms Huh [41] and Salloum [43] and achieves comparable performance to [40] in terms of F 1 -score.…”

Section: ) Comparison With Other State-of-the-art Splicing Localizatmentioning

confidence: 99%

See 2 more Smart Citations

Deep Learning Local Descriptor for Image Splicing Detection and Localization

Rao

Zhao

2020

IEEE Access

View full text Add to dashboard Cite

In this paper, a novel image splicing detection and localization scheme is proposed based on the local feature descriptor which is learned by deep convolutional neural network (CNN). A two-branch CNN, which serves as an expressive local descriptor is presented and applied to automatically learn hierarchical representations from the input RGB color or grayscale test images. The first layer of the proposed CNN model is used to suppress the effects of image contents and extract the diverse and expressive residual features, which is deliberately designed for image splicing detection applications. In specific, the kernels of the first convolutional layer are initialized with an optimized combination of the 30 linear high-pass filters used in calculation of residual maps in spatial rich model (SRM), and is fine-tuned through a constrained learning strategy to retain the high-pass filtering properties for the learned kernels. Both the contrastive loss and cross entropy loss are utilized to jointly improve the generalization ability of the proposed CNN model. With the block-wise dense features for a test image extracted by the pre-trained CNN-based local descriptor, an effective feature fusion strategy, known as block pooling, is adopted to obtain the final discriminative features for image splicing detection with SVM. Based on the pre-trained CNN model, an image splicing localization scheme is further developed by incorporating the fully connected conditional random field (CRF). Extensive experimental results on several public datasets show that the proposed CNN based scheme outperforms some state-of-the-art methods not only in image splicing detection and localization performance, but also in robustness against JPEG compression. INDEX TERMS Image splicing detection, splicing localization, convolutional neural network, feature fusion, conditional random field (CRF).

show abstract

Section: ) Comparison With Other State-of-the-art Splicing Localizatmentioning

confidence: 99%

Section: ) Comparison With Other State-of-the-art Splicing Localizatmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning Local Descriptor for Image Splicing Detection and Localization

Rao

Zhao

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…copy-move [1], [21], seam carving [22]. Despite the tremendous progress so far, much potential and many more discoveries lie ahead because of the breakthrough in deep learning [23], [24], many CNN-based methods are investigated and achieve significant improvements [2], [5], [9], [25]. However, as we describe in Section I, these methods all investigate individual images and can not provide the source of splicing images.…”

Section: Related Workmentioning

confidence: 99%

Constrained Image Splicing Detection and Localization With Attention-Aware Encoder-Decoder and Atrous Convolution

Liu

Zhao

2020

IEEE Access

View full text Add to dashboard Cite

Constrained image splicing detection and localization (CISDL) is a newly formulated image forensics task and plays an important role in verifying the generating process of a forged image. CISDL conducts dense matching between two investigated images and detects whether one image has forged regions pasted from the other. In this work, we introduce a novel attention-aware encoder-decoder deep matching network named as AttentionDM for CISDL. An encoder-decoder with atrous convolution is newly designed for hierarchical features dense matching and fine-grained masks generation. A novel attention-aware correlation computation module is built on normalization operations and informative features recalibration with channel attention blocks. Last but not least, VGG and ResNets are respectively formulated as feature extractors for comprehensive comparisons in CISDL. Extensive experiments demonstrate the superior performance of AttentionDM over the state-of-the-art methods.

show abstract

“…However, the time complexity of existing methods is high, especially in the feature matching stage, and the location of tampered regions is not accurate enough to meet the practical requirements. Since in practical forensics applications, figuring out the tampered regions compared to forgery detections is more important and necessary [33]. In this paper, an improved CMFD method is proposed, which includes clustering the keypoints before matching and locating the tampered regions at pixel level.…”

Section: Introductionmentioning

confidence: 99%

Copy-Move Forgery Detection Based on Keypoint Clustering and Similar Neighborhood Search Algorithm

2020

View full text Add to dashboard Cite

Copy-move is one of the most commonly used methods of tampering with digital images. Keypoint-based detection is recognized as effective in copy-move forgery detection (CMFD). This paper proposes an efficient CMFD method via clustering SIFT keypoints and searching the similar neighborhoods to locate tampered regions. In the proposed method, the keypoints are clustered based on scale and color, grouped into several smaller clusters and matched separately, which reduce the high time complexity caused in matching caused by the high dimensionality of SIFT. In order to locate the tampered regions accurately at pixel level finally, a novel localization algorithm is designed to compare the similar neighborhoods of matching pairs by two similarity measures, and mark the tampered regions in pixels iteratively. We experimented on three different image data sets including kinds of tampering means to compare and verify the effectiveness and robustness of proposed method. The experimental results show that the proposed method is superior to existing state-of-art methods in terms of matching time complexity, detection reliability and forgery location accuracy. INDEX TERMS Copy-move forgery detection, digital image forensics, keypoint clustering, similar neighborhood search algorithm.

show abstract

Image Manipulation Detection and Localization Based on the Dual-Domain Convolutional Neural Networks

Cited by 36 publications

References 26 publications

Deep Learning Local Descriptor for Image Splicing Detection and Localization

Deep Learning Local Descriptor for Image Splicing Detection and Localization

Constrained Image Splicing Detection and Localization With Attention-Aware Encoder-Decoder and Atrous Convolution

Copy-Move Forgery Detection Based on Keypoint Clustering and Similar Neighborhood Search Algorithm

Contact Info

Product

Resources

About