A fast technique to detect copy-move image forgery with reflection and non-affine transformation attacks

作者：

Highlights：

• Proposed method is robust against various complex and novel attacks applied over images.

• Utilizes prominent techniques, namely CenSurE, LIPID, and Fuzzy C-Means clustering.

• Has ability to detect tampered images formulated by manipulation over images collected from versatile datasets.

• A fast methodology as compared to state-of-the-art methods.

摘要

•Proposed method is robust against various complex and novel attacks applied over images.•Utilizes prominent techniques, namely CenSurE, LIPID, and Fuzzy C-Means clustering.•Has ability to detect tampered images formulated by manipulation over images collected from versatile datasets.•A fast methodology as compared to state-of-the-art methods.

论文关键词：Center Surround Extrema detector,Copy-move image forgery,Digital image forensics,Fuzzy C-Means clustering,Local Image Permutation Interval Descriptor,Passive authentication

论文评审过程：Received 7 June 2020, Revised 30 March 2021, Accepted 22 May 2021, Available online 1 June 2021, Version of Record 6 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115282