Enhancing Hyperspectral Image Classification With Spectral–Spatial Convolutional FourierKAN Transformer

The dominance of transformer architectures in hyperspectral image (HSI) classification, driven by their self-attention mechanisms and multilayer perceptron (MLP), faces critical challenges in mathematical interpretability and spectral redundancy sensitivity. To address these limitations, we propose...

Descripción completa

Detalles Bibliográficos

Autores principales:	Zixuan Xue, Zhijing Ye, Boyu Wang, Yuqing Zhang, Chengyong Zheng, Jiangtao Peng
Formato:	Artigo
Lenguaje:	Inglês
Publicado:	IEEE 2026-01-01
Colección:	IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Materias:	Hyperspectral image (HSI) classification Kolmogorov–Arnold network (KAN) multilayer perceptron (MLP) transformer
Acceso en línea:	https://ieeexplore.ieee.org/document/11319177/
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!