Cost-sensitive stacked sparse auto-encoder models to detect striped stem borer infestation on rice based on hyperspectral imaging

作者：

Highlights：

• Cost-sensitive stacked sparse auto-encoder model was proposed to detect the striped stem borer infestation on rice.

• The cost function of SSAE was improved to become sensitive to early infestation stage.

• The optimal structure of cost-sensitive SSAE was developed by multiple structures.

• Cost-sensitive SSAE was superior in variables reduction than other classical feature extraction and selection methods.

摘要

•Cost-sensitive stacked sparse auto-encoder model was proposed to detect the striped stem borer infestation on rice.•The cost function of SSAE was improved to become sensitive to early infestation stage.•The optimal structure of cost-sensitive SSAE was developed by multiple structures.•Cost-sensitive SSAE was superior in variables reduction than other classical feature extraction and selection methods.

论文关键词：Striped stem borer,Hyperspectral imaging technology,Stacked sparse auto-encoder,Cost-sensitive function,Early detection

论文评审过程：Received 6 July 2018, Revised 7 December 2018, Accepted 2 January 2019, Available online 14 January 2019, Version of Record 15 February 2019.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.01.003