Article · IEEE Transactions on Geoscience and Remote Sensing · January 1, 2026

Concept Disentangled Transformer for Hyperspectral Image Classification

Muhammad Imran Ahmad, Manuel Mazzara, Gemine Vivone, Danfeng Hong

Abstract

Hyperspectral image (HSI) classification remains challenging due to high spectral dimensionality, extremely limited labeled samples, and complex spatial–spectral dependencies. Recent Transformer-based models provide strong contextual modeling, yet their high-dimensional latent representations are often weakly structured under low-label regimes, which can lead to unstable class separation and limited analyzability of the decision process. This article proposes CDFormer (CDF), a Transformer framework that introduces an explicit concept bottleneck (CD) to impose a structured intermediate representation for classification. CDF projects attention-enhanced token features into a low-dimensional concept space composed of structured latent coordinates for decision decomposition and predicts labels through a linear class-concept readout, yielding an exact logit decomposition into concept-wise contributions rather than predefined semantic attributes. To improve representation learning under scarce supervision, we train CDF using a composite objective that combines standard classification loss with orthogonality and sparsity regularization (SP) and a supervised contrastive term applied in concept space, encouraging reduced feature redundancy, selective coordinate activation, and improved intraclass compactness/interclass separation. Experiments on five benchmark HSI datasets demonstrate that CDF achieves consistent performance compared with recent state-of-the-art baselines.

Remote-Sensing Image ClassificationAdvanced Neural Network ApplicationsSpectroscopy and Chemometric Analyseshyperspectral image classificationtransformer modelsconceptual space

Concept Disentangled Transformer for Hyperspectral Image Classification

Abstract

More by Manuel Mazzara