Recaptured image forensics based on local ternary count of high order prediction error

作者：

Highlights：

• A novel recaptured image forensics method based on local ternary count of high order prediction error is proposed.

• An efficient calculation algorithm of high order prediction error maps based on linear prediction is proposed.

• Experiments performed on three databases demonstrate this method has good performance in both detection accuracy and generalization ability.

摘要

•A novel recaptured image forensics method based on local ternary count of high order prediction error is proposed.•An efficient calculation algorithm of high order prediction error maps based on linear prediction is proposed.•Experiments performed on three databases demonstrate this method has good performance in both detection accuracy and generalization ability.

论文关键词：Image forensics,Recaptured image detection,Recaptured tampered image,Image credibility,Prediction error,Local ternary count

论文评审过程：Received 17 August 2020, Revised 21 January 2022, Accepted 13 February 2022, Available online 18 February 2022, Version of Record 5 March 2022.

论文官网地址：https://doi.org/10.1016/j.image.2022.116662