Automated Image Splicing Detection from Noise Estimation in Raw Images

Abstract : Splicing is a common image manipulation technique in which a region from a first image is pasted onto a second image to alter its content. In this paper, we use the fact that different images have different noise characteristics, according to the camera and lighting conditions during the image acquisition. The proposed method automatically detects image splicing in raw images by highlighting local noise inconsistencies within a quadtree scan of the image. The image noise is modelized by both Gaussian and Poisson noise components. We demonstrate the efficiency and robustness of our method on several images generated with an automated image splicing.
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal-upec-upem.archives-ouvertes.fr/hal-01510075
Contributor : Vincent Nozick <>
Submitted on : Wednesday, April 19, 2017 - 4:42:53 AM
Last modification on : Thursday, July 5, 2018 - 2:26:41 PM

File

julliand_ICDP_2015.pdf
Files produced by the author(s)

Identifiers

Citation

Thibaut Julliand, Vincent Nozick, Hugues Talbot. Automated Image Splicing Detection from Noise Estimation in Raw Images. 6th International Conference on Imaging for Crime Prevention and Detection, Jul 2015, London, United Kingdom. pp.13-18, ⟨10.1049/ic.2015.0111⟩. ⟨hal-01510075⟩

Share

Metrics

Record views

180

Files downloads

203