Source printer identification using printer specific pooling of letter descriptors

作者：

Highlights：

• Extensive analysis of printer signature variations across a printed document.

• Introduce a location-based pooling technique for source printer identification.

• Proposed system works with a correlation-based prediction method.

• Proposed system performs better than existing methods under cross-font scenario.

摘要

•Extensive analysis of printer signature variations across a printed document.•Introduce a location-based pooling technique for source printer identification.•Proposed system works with a correlation-based prediction method.•Proposed system performs better than existing methods under cross-font scenario.

论文关键词：Printer classification,Printer forensics,Source identification,Multimedia forensics,Local binary pattern,Local texture descriptor

论文评审过程：Received 30 November 2020, Revised 21 September 2021, Accepted 27 November 2021, Available online 22 December 2021, Version of Record 24 December 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116344