TY - GEN
T1 - ENERGY-BASED GENERATIVE MODELS WITH MORPHOLOGICAL ATTENTION NETWORKS FOR HYPERSPECTRAL IMAGE CLASSIFICATION
T2 - 32nd IEEE International Conference on Image Processing, ICIP 2025
AU - Hossain, Mohammad Arif
AU - Sarker, Yeahia
AU - Liu, Weiqi
AU - Ansari, Nirwan
N1 - Publisher Copyright:
©2025 IEEE.
PY - 2025
Y1 - 2025
N2 - This paper presents an innovative framework that synergistically integrates energy-based generative models with morphological attention networks for robust hyperspectral image classification. Despite recent advances in data-driven approaches, existing methods struggle to model complex data distributions while preserving geometric invariance and maintaining stable representations under limited supervision. To address these challenges, we propose a novel architecture that combines three key components: (1) a morphological attention module that fuses learned structural elements with multi-head attention through parallel dilation and erosion pathways, (2) an uncertainty-aware energy-based learning paradigm that employs adaptive margin constraints and Langevin dynamics sampling, and (3) a multi-scale feature extraction block with parallel convolutional branches for dynamic spectral-spatial feature integration. Extensive experimental validation on benchmark datasets demonstrates our framework’s superior ability over state-of-the-art methods.
AB - This paper presents an innovative framework that synergistically integrates energy-based generative models with morphological attention networks for robust hyperspectral image classification. Despite recent advances in data-driven approaches, existing methods struggle to model complex data distributions while preserving geometric invariance and maintaining stable representations under limited supervision. To address these challenges, we propose a novel architecture that combines three key components: (1) a morphological attention module that fuses learned structural elements with multi-head attention through parallel dilation and erosion pathways, (2) an uncertainty-aware energy-based learning paradigm that employs adaptive margin constraints and Langevin dynamics sampling, and (3) a multi-scale feature extraction block with parallel convolutional branches for dynamic spectral-spatial feature integration. Extensive experimental validation on benchmark datasets demonstrates our framework’s superior ability over state-of-the-art methods.
KW - Attention Mechanism
KW - Generative Models
KW - Hyperspectral Image Classification
KW - Morphological Operation
UR - https://www.scopus.com/pages/publications/105028624417
UR - https://www.scopus.com/pages/publications/105028624417#tab=citedBy
U2 - 10.1109/ICIP55913.2025.11084715
DO - 10.1109/ICIP55913.2025.11084715
M3 - Conference contribution
AN - SCOPUS:105028624417
T3 - Proceedings - International Conference on Image Processing, ICIP
SP - 2282
EP - 2287
BT - 2025 IEEE International Conference on Image Processing, ICIP 2025 - Proceedings
PB - IEEE Computer Society
Y2 - 14 September 2025 through 17 September 2025
ER -