Corn (No Till) (a) (b) (c) (d) (e) (f) (g) (h) Corn (Minimum Till) Grass/ Pasture Hay-Windowed Soybean (No Till) Soybean (Minimum Till) Soybean (Clean Till) Woods FIGURE 7. The thematic maps resulting from dimensionality-reduction methods for the Indian Pines data set, with eight classes: (a) LDA: 74.74%. (b) LFDA: 80.97%. (c) SGDA: 81.18%. (d) CGDA: 84.35%. (e) KDA: 81.74%. (f) KLFDA: 86.65%. (g) KSGDA: 85.84%. (h) KCGDA: 86.50%. (a) (b) (c) (d) (e) (f) (g) (h) Unlabeled Metal Sheets Asphalt Bare Soil Meadows Bitumen Gravel Bricks Trees Shadows FIGURE 8. The thematic maps resulting from dimensionality-reduction methods for the University of Pavia data set, with nine classes: (a) LDA: 88.74%. (b) LFDA: 92.43%. (c) SGDA: 93.92%. (d) CGDA: 93.06%. (e) KDA: 90.36%. (f) KLFDA: 92.59%. (g) KSGDA: 93.83%. (h) KCGDA: 93.62%. march 2018 ieee Geoscience and remote sensing magazine 27