Non-separable rotation moment invariants
作者：
Highlights：
• New invariants to image rotation are presented.
• The invariants are based on non-separable moments, which is a difference from all previously published invariants.
• Non-separability of the moments contributes to better recognition power and to higher robustness.
• The invariants can be efficiently evaluated by recurrent relations.
• The experiments compare the new invariants to three other state-of-the art ones.
摘要
•New invariants to image rotation are presented.•The invariants are based on non-separable moments, which is a difference from all previously published invariants.•Non-separability of the moments contributes to better recognition power and to higher robustness.•The invariants can be efficiently evaluated by recurrent relations.•The experiments compare the new invariants to three other state-of-the art ones.
论文关键词：Image recognition,Rotation invariants,Non-separable moments,Appell polynomials,Bi-orthogonality,Recurrent relation
论文评审过程：Received 10 January 2021, Revised 16 February 2022, Accepted 21 February 2022, Available online 22 February 2022, Version of Record 2 March 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108607
