Blind image clustering based on the Normalized Cuts criterion for camera identification

作者：

Highlights：

• Images are clustered according to their acquisition source.

• Clustering is based on Normalized Cuts criterion.

• Performances are better than other state-of-the-art methods.

摘要

Highlights•Images are clustered according to their acquisition source.•Clustering is based on Normalized Cuts criterion.•Performances are better than other state-of-the-art methods.

论文关键词：Image forensics,Source identification,Clustering,Normalized Cuts

论文评审过程：Received 18 December 2013, Revised 18 July 2014, Accepted 18 July 2014, Available online 30 July 2014.

论文官网地址：https://doi.org/10.1016/j.image.2014.07.003