Low-rank sparse feature selection for image classification

作者：

Highlights：

• The LRSSRcan reduce the negativeeffect of noiseand redundant features.

• The HCcan solve the problem of overfittingand improve the classification accuracy.

• The LRSSRcan be widely applied in general image feature selection.

摘要

•The LRSSRcan reduce the negativeeffect of noiseand redundant features.•The HCcan solve the problem of overfittingand improve the classification accuracy.•The LRSSRcan be widely applied in general image feature selection.

论文关键词：Feature selection,SVM,Convolutional neural network

论文评审过程：Received 13 March 2021, Revised 28 May 2021, Accepted 27 July 2021, Available online 24 October 2021, Version of Record 30 October 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115685