# Hyperspectral Image Classification on WHU-Hi-HanChuan (1% train) Source: [Hyperspectral Image Classification on WHU-Hi-HanChuan (1% train)]() Collection index: [llms.txt]() Data updated: 2026-04-28T08:50:51.874Z Evaluation of Hyperspectral Image Classification performance on the WHU-Hi-HanChuan dataset using only 1% of the training data. Metrics include Overall Accuracy (OA), Average Accuracy (AA), Kappa coefficient, and per-class accuracies for 16 classes. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Hyperspectral Image Classification | | Dataset (including any stated split) | WHU-Hi-HanChuan (1% train) | | Metric | Per-Class Accuracy 1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Per-Class Accuracy 1](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 2](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 3](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 4](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 5](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 6](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 7](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 8](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 9](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 10](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 11](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 12](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 13](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 14](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 15](): Higher is better; unit: PERCENT. - [Per-Class Accuracy 16](): Higher is better; unit: PERCENT. - [Overall Accuracy (OA)](): Higher is better; unit: PERCENT. - [Average Accuracy (AA)](): Higher is better; unit: PERCENT. - [Kappa Coefficient](): Higher is better; unit: PERCENT. ## Recorded results | Method | Per-Class Accuracy 1 | Result date | Report paper | | --- | --- | --- | --- | | SCT-Net | 99.56 | Not recorded | 69f03f4c55a3c32af89fa2df | | HybridSN | 99.26 | Not recorded | 69f03f4c55a3c32af89fa2df | | MASSFormer | 99.01 | Not recorded | 69f03f4c55a3c32af89fa2df | | SSFTT | 98.49 | Not recorded | 69f03f4c55a3c32af89fa2df | | SVM | 95.95 | Not recorded | 69f03f4c55a3c32af89fa2df | | 2D-CNN | 95.53 | Not recorded | 69f03f4c55a3c32af89fa2df | | 3D-CNN | 95.19 | Not recorded | 69f03f4c55a3c32af89fa2df | | SpectralFormer | 94.4 | Not recorded | 69f03f4c55a3c32af89fa2df | | SS-Mamba | 92.13 | Not recorded | 69f03f4c55a3c32af89fa2df | ### SCT-Net Per-Class Accuracy 1: 99.56. Result ID: 69f0749e47f34ce924d67711. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### HybridSN Per-Class Accuracy 1: 99.26. Result ID: 69f0749e47f34ce924d6770c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### MASSFormer Per-Class Accuracy 1: 99.01. Result ID: 69f0749e47f34ce924d6770f. [Reporting paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### SSFTT Per-Class Accuracy 1: 98.49. Result ID: 69f0749e47f34ce924d6770e. [Reporting paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### SVM Per-Class Accuracy 1: 95.95. Result ID: 69f0749e47f34ce924d67709. [Reporting paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### 2D-CNN Per-Class Accuracy 1: 95.53. Result ID: 69f0749e47f34ce924d6770a. [Reporting paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### 3D-CNN Per-Class Accuracy 1: 95.19. Result ID: 69f0749e47f34ce924d6770b. [Reporting paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### SpectralFormer Per-Class Accuracy 1: 94.4. Result ID: 69f0749e47f34ce924d6770d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. ### SS-Mamba Per-Class Accuracy 1: 92.13. Result ID: 69f0749e47f34ce924d67710. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training sample ratio":"1%"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.