Image splicing localization based on CFA-artifacts analysis

Varlamova, A.; Kuznetsov, A.

doi:10.18287/2412-6179-2017-41-6-920-930

Cited by 7 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Commonly, the tampered images that come from the Internet involve economic, political, media, military, technology, medical, and judicial fields [Wei, Wang and Ma (2017)] and thus can give a very bad influence on the country and society. In this case, the image splicing detection techniques are of significant scientific importance, but they are traditionally concentrated on using the methods of pattern noise [Siwei, Xunyu and Xing (2014); Yao, Wang, Zhang et al (2017)], color filter array (CFA) [Ferrara, Bianchi, De Rosa et al (2012); Varlamova and Kuznetsov (2017)] and blocking [Bianchi and Piva (2012), Bianchi, De Rosa and Piva (2011)] to detect, which have some drawbacks because they need some prior information and can only handle a certain type of forgery. Moreover, the current image splicing detection technologies can only solve whether an image has undergone splicing, without the ability to locate the tampered region.…”

Section: Introductionmentioning

confidence: 99%

Improved Fully Convolutional Network for Digital Image Region Forgery Detection

Zhang¹,

Li²,

Niu³

et al. 2019

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

With the rapid development of image editing techniques, the image splicing behavior, typically for those that involve copying a portion from one original image into another targeted image, has become one of the most prevalent challenges in our society. The existing algorithms relying on hand-crafted features can be used to detect image splicing but unfortunately lack precise location information of the tampered region. On the basis of changing the classifications of fully convolutional network (FCN), here we proposed an improved FCN that enables locating the spliced region. Specifically, we first insert the original images into the training dataset that contains tampered images forming positive and negative samples and then set the ground truth masks of the original images to be black images. The purpose of forming positive and negative samples is to guide the improved FCN to distinguish the differences between the original images and spliced images. After these steps, we conducted an experiment to verify our proposal, and the results reveal that the improved FCN really can locate the spliced region. In addition, the improved FCN achieves improved performance compared to the already-existing algorithms, thereby providing a feasible approach for digital image region forgery detection.

show abstract