Hyperspectral remote sensing IQA via learning multiple kernels from mid-level features

作者：

Highlights：

• The designed mid-level features can better represent attributes of hyperspectral images.

• The designed framework can integrate multiple scale features of hyperspectral images.

• Comprehensive experiments have demonstrated the effectiveness of our proposed HIQA algorithm.

摘要

•The designed mid-level features can better represent attributes of hyperspectral images.•The designed framework can integrate multiple scale features of hyperspectral images.•Comprehensive experiments have demonstrated the effectiveness of our proposed HIQA algorithm.

论文关键词：Hyperspectral image quality assessment,Mid-level feature,Deep features,Multiple kernel learning,Quality-aware

论文评审过程：Received 29 July 2019, Revised 8 January 2020, Accepted 17 January 2020, Available online 24 January 2020, Version of Record 24 February 2020.

论文官网地址：https://doi.org/10.1016/j.image.2020.115804