Fig 1.
Schematic diagram of SSGIE-KFCM algorithm flow.
Fig 2.
Schematic diagram of the improved FA process.
Fig 3.
Flow diagram of FA-FCM algorithm.
Table 1.
Source code content of SSGIE-KFCM algorithm.
Table 2.
Experimental environmental conditions.
Table 3.
Sensitivity analysis of hyperparameters.
Fig 4.
Training loss and testing loss results of the fusion algorithm.
Table 4.
Performance comparison of different hybrid clustering band selection methods.
Fig 5.
Classification accuracy of band subsets-Pavia University dataset.
Fig 6.
RS image band classification accuracy.
Fig 7.
Classification accuracy values under different sample testing ratios.
Table 5.
Classification confusion outcomes of different approaches.
Table 6.
Detection of different RS image bands in the Indian Pines dataset.
Table 7.
Detection of different RS image bands in the Pavia University dataset.
Fig 8.
Performance of different ground feature band selection.
Table 8.
Performance evaluation of band selection for different algorithms (with 50 bands).
Table 9.
Calculation time of spectral features in the final classification stages (s).
Table 10.
Comparison results of calculation time for research methods (s).
Table 11.
McNemar test results (p-values) of SGIE-KFCM and other methods.
Table 12.
Theoretical computational complexity comparison and numerical estimation.
Table 13.
Performance comparison of various methods when selecting 30 bands on the Salinas dataset.
Table 14.
The ablation research results on the Pavia University dataset.