Geometric invariant features in the Radon transform domain for near-duplicate image detection

作者：

Highlights：

• A geometric invariant feature model in Radon transform domain is proposed.

• Four kinds of geometric invariant features are developed based on the model.

• A uniform sampling technique is presented to combine different invariant features.

• The reported features perform better than others in near-duplicate image detection.

摘要

Highlights•A geometric invariant feature model in Radon transform domain is proposed.•Four kinds of geometric invariant features are developed based on the model.•A uniform sampling technique is presented to combine different invariant features.•The reported features perform better than others in near-duplicate image detection.

论文关键词：Enhanced Radon feature,(ERF),Moment pattern,(MP),High-order invariant moment,(HOIM),Integral of Fourier transform,(IOFT),Arc length and area ratio,(ALAR),Geometric invariants,Radon transform,Near-duplicate image detection

论文评审过程：Received 9 September 2012, Revised 24 October 2013, Accepted 12 May 2014, Available online 21 May 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.05.009