ABSTRACT
The architectures based on Multi-Layer Perceptron (MLP) have attracted great attention in hyperspectral image (HSI) classification recently, due to their simplified and efficient architectures. However, such architectures are qualified by the rigid positional relationships between weights and feature elements, inhibiting their capacity to effectively extract diversified features. To address these challenges, An adaptive spatial-shift MLP (AS2MLP) is presented to dynamically modify spatial features by parameterizing learnable spatial offsets. In this way, the AS2MLP can facilitate sample-specific spatial shifts, aligning spatial structures more effectively. Then, An innovative adaptive spatial-shift block (AS2block) is designed to adaptively shift spatial features along distinct spatial axes, enabling the extraction of diversified features separately. It also implements a re-weighting strategy to mitigate redundant features. Building on this foundation, the proposed adaptive spatial-shift network (AS2Net) is for HSI classification. The dual-path AS2Net employs AS2blocks and MLPs for channel mixing, facilitating an adaptive integration of dynamic spatial contextual information dispersed across a range of spectra. The effectiveness of this model is demonstrated using five widely used HSI datasets.
Acknowledgements
The authors would like to express their gratitude to the reviewers for their insightful remarks and ideas on how to improve the paper’s quality.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Authors’ contributions
Qiaoyu Ma: Conceptualization, Methodology, Writing – Review & Editing.
Heng Zhou: Supervision, Methodology, Writing – Original Draft.
Yanan Jiang: Analysis, Visualization, Writing – Editing.
Zitong Zhang: Data Collection, Investigation, Writing – Editing.
Chunlei Zhang: Project Administration, Writing – Review.
Ethical consideration and informed consent
During the whole research process, all authors followed international guidelines and ensured that the animals were not harmed. No private or personally identifiable information was utilized. The datasets employed in this study were publicly available and were accessed and used in compliance with the terms and conditions under which they were released. We acknowledge the data providers and have ensured that our usage aligns with any licences or permissions associated with the datasets. No authors have been omitted and all individuals included as author in the manuscript meet all of the above requirements.
Notes
1. The data that support the findings of this study are openly available at.
https://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes#Pavia_University_scene
2. The data that support the findings of this study are openly available at.