Identification of chrysanthemum using hyperspectral imaging based on few-shot class incremental learning

Zeyi Cai, Mengyu He, Cheng Li, Hengnian Qi, Ruibin Bai, Jian Yang, Chu Zhang

Abstract

Highlights•Origin and variety identification of Chrysanthemum were studied by NIR-HSI.•Advanced deep learning and visualization methods were utilized for classification.•Few-shot Class-Incremental learning using the Replay training strategy was studied.•Good performances were obtained by Few-shot Class-Incremental learning approach.
BibTeX
@article{cai2023identification,
  author = {Zeyi Cai and Mengyu He and Cheng Li and Hengnian Qi and Ruibin Bai and Jian Yang and Chu Zhang},
  title = {{Identification of chrysanthemum using hyperspectral imaging based on few-shot class incremental learning}},
  journal = {Comput. Electron. Agric.},
  year = {2023},
  doi = {10.1016/j.compag.2023.108371},
}

Loading related content...

Figure