Multiple-rank supervised canonical correlation analysis for feature extraction, fusion and recognition

作者：

Highlights：

• A supervised two-dimensional CCA method is proposed for feature extraction.

• By further exploration another supervised framework is also presented.

• Experimental results of our methods superior/comparable to the literature.

摘要

•A supervised two-dimensional CCA method is proposed for feature extraction.•By further exploration another supervised framework is also presented.•Experimental results of our methods superior/comparable to the literature.

论文关键词：Two-dimensional canonical correlation analysis (2D-CCA),Supervised information,Feature extraction,Feature fusion,Pattern recognition

论文评审过程：Received 20 November 2016, Revised 16 April 2017, Accepted 6 May 2017, Available online 8 May 2017, Version of Record 12 May 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.05.017