Adaptive Band Selection for Hyperspectral Classification with Spatially Disjoint Evaluation
Quick Answer
The SGBR-HC method improves hyperspectral band selection by using a two-stage process for training sparse gates, achieving the highest mean overall accuracy and Cohen's kappa with about twenty bands on Pavia University and Houston 2013 datasets.
Quick Take
The SGBR-HC method improves hyperspectral band selection by using a two-stage process for training sparse gates, achieving the highest mean overall accuracy and Cohen's kappa with about twenty bands on Pavia University and Houston 2013 datasets. Bypassing the initial ranking stage significantly reduces accuracy, highlighting its importance in the evaluation process.
Key Points
- SGBR-HC uses supervised spectral ranking for initializing trainable sparse gates.
- Achieved highest mean overall accuracy and Cohen's kappa with approximately twenty bands.
- Bypassing Stage-1 reduces overall accuracy by 8.84 pp on Pavia University.
- Random pixel splits inflate overall accuracy by 30.56 pp, indicating spatial leakage issues.
- Method verified through retraining a fresh classifier on selected bands.
Article Content
From source RSS / original summaryarXiv:2606. 06684v1 Announce Type: new Abstract: Hyperspectral band selection methods based on differentiable selectors can be sensitive to initialization and to extracting a final discrete subset, while prescribed band counts limit flexibility.
We propose SGBR-HC (Spectral-Group Band Ranking with Hard-Concrete initialization), a two-stage method that uses a supervised spectral ranking to initialize trainable sparse gates rather than treating ranking as a fixed selection rule, letting the number of selected bands be determined by training. Stage-1 scores candidate bands from training pixels by class discriminability and spectral diversity; this ranking seeds the gate logits for Stage-2, which trains the sparse gates jointly with a spatial classifier.
Under spatially disjoint evaluation on Pavia University and Houston 2013, verified by retraining a fresh classifier on the selected bands, SGBR-HC achieves the highest mean overall accuracy and Cohen's kappa with approximately twenty bands. Bypassing Stage-1 degrades OA by 8. 84 pp on Pavia University and 22. 15 pp on Houston 2013, confirming the ranking prior's role. Random pixel splits inflate OA on Pavia University by 30. 56 pp, underscoring spatial leakage as a critical evaluation confound.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.