Skip to Main content Skip to Navigation
Conference papers

Using Neural Networks for Fake Colorized Image Detection

Abstract : Modern colorization techniques can create artificially-colorized images that are indistinguishable from natural color images. As a result, the detection of fake colorized images is attracting the interest of the digital forensics research community. This chapter tackles the challenge by introducing a detection approach that leverages neural networks. It analyzes the statistical differences between fake colorized images and their corresponding natural images, and shows that significant differences exist. A simple, but effective, feature extraction technique is proposed that utilizes cosine similarity to measure the overall similarity of normalized histogram distributions of various channels for natural and fake images. A special neural network with a simple structure but good performance is trained to detect fake colorized images. Experiments with datasets containing fake colorized images generated by three state-of-the-art colorization techniques demonstrate the performance and robustness of the proposed approach.
Document type :
Conference papers
Complete list of metadata

Cited literature [22 references]  Display  Hide  Download
Contributor : Hal Ifip Connect in order to contact the contributor
Submitted on : Tuesday, April 7, 2020 - 10:37:14 AM
Last modification on : Tuesday, April 7, 2020 - 10:42:38 AM


Files produced by the author(s)


Distributed under a Creative Commons Attribution 4.0 International License



Yuze Li, Yaping Zhang, Liangfu Lu, Yongheng Jia, Jingcheng Liu. Using Neural Networks for Fake Colorized Image Detection. 15th IFIP International Conference on Digital Forensics (DigitalForensics), Jan 2019, Orlando, FL, United States. pp.201-215, ⟨10.1007/978-3-030-28752-8_11⟩. ⟨hal-02534604⟩



Record views


Files downloads