Laser Journal, Volume. 46, Issue 3, 98(2025)

A lifelong learning method for hyperspectral image classification using multi-level knowledge distillation

JIANG Zihui, LI Zhaokui*, and WANG Ke
Author Affiliations
  • School of Computer Science, Shenyang Aerospace University, Shenyang 110136, China
  • show less

    To address the catastrophic forgetting problem in lifelong learning, this paper proposes a lifelong learning method for hyperspectral image classification using multi-level knowledge distillation. Initially, a feature extractor based on multi-modal alignment is designed to fully leverage the spatial-spectral information and label text information of hyperspectral images. Additionally, a multi-level knowledge distillation strategy is devised to effectively preserve the multi-modal knowledge from previous phases. The proposed method was experimented on two public hyperspectral datasets. Compared to current state-of-the-art methods, the proposed method showed an average accuracy improvement of 15%-18% on the Pavia University dataset and 1%-8% on the Botswana dataset.

    Tools

    Get Citation

    Copy Citation Text

    JIANG Zihui, LI Zhaokui, WANG Ke. A lifelong learning method for hyperspectral image classification using multi-level knowledge distillation[J]. Laser Journal, 2025, 46(3): 98

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Oct. 24, 2024

    Accepted: Jun. 12, 2025

    Published Online: Jun. 12, 2025

    The Author Email: LI Zhaokui (lzk@sau.edu.cn)

    DOI:10.14016/j.cnki.jgzz.2025.03.098

    Topics