Acta Optica Sinica, Volume. 38, Issue 4, 0428001(2018)
A Semi-Supervised Dimension Reduction Method for Polarimetric SAR Image Classification
Figures & Tables(17)
Fig. 1. Spatial scatter distribution of polarimetric features of Flevoland. (a) σHH0, σHV0, σVV0; (b) H, α, A; (c) ρHH-VV, ρHH-HV, ρVV-HV
Fig. 3. (a) Pseudocolor image; (b) labeled truth in classification of RADARSAT-2 Flevoland
Fig. 4. (a) Pseudocolor image; (b) labeled truth in classification of AIRSAR Flevoland 1
Fig. 5. (a) Pseudocolor image; (b) labeled truth in classification of AIRSAR Flevoland 2
Fig. 6. Relationship between training data number and classification accuracy. (a) RADARSAT-2 Flevoland; (b) AIRSAR Flevoland 1; (c) AIRSAR Flevoland 2
Fig. 7. Relationship between dimension number and classification accuracy. (a) RADARSAT-2 Flevoland; (b) AIRSAR Flevoland 1; (c) AIRSAR Flevoland 2
Fig. 8. Relationship between labeled sample proportion and classification accuracy. (a) RADARSAT-2 Flevoland; (b) AIRSAR Flevoland 1; (c) AIRSAR Flevoland 2
Fig. 9. Relationship between balance factor and classification accuracy. (a) RADARSAT-2 Flevoland; (b) AIRSAR Flevoland 1; (c) AIRSAR Flevoland 2
Fig. 11. Classification maps of RADARSAT-2 Flevoland. (a) SLDA+KNN; (b) KNN; (c) PCA+KNN; (d) LPP+KNN; (e) LDA+KNN; (f) local detail maps
Fig. 12. Classification maps of AIRSAR Flevoland 1. (a) SLDA+KNN; (b) KNN; (c) PCA+KNN; (d) LPP+KNN; (e) LDA+KNN
Fig. 13. 3D scatter distribution diagram of AIRSAR Flevoland 2. (a) SLDA; (b) PCA; (c) LPP; (d) LDA
Fig. 14. Classification maps of AIRSAR Flevoland 2. (a) SLDA+KNN; (b) KNN; (c) PCA+KNN; (d) LPP+KNN; (e) LDA+KNN
Table 1. Classification accuracy of RADARSAT-2 Flevoland%
View tableTable 1. Classification accuracy of RADARSAT-2 Flevoland%
Category SLDA PCA LPP KNN LDA Water 93.07 93.77 93.57 93.81 94.18 Farmland 91.94 88.15 86.50 88.94 91.13 Forest 88.39 81.59 81.68 83.01 81.05 Building 81.02 68.65 71.33 67.42 69.10 OA 88.88 83.24 83.23 83.67 84.25
Table 2. Classification accuracy of AIRSAR Flevoland 1%
View tableTable 2. Classification accuracy of AIRSAR Flevoland 1%
Category SLDA PCA LPP KNN LDA Steam bean 96.88 94.99 9.70 94.30 98.37 Forest 98.59 95.03 47.01 95.08 95.66 Potato 89.77 84.90 23.03 85.23 88.04 Lucerne 92.74 90.70 7.34 90.82 92.99 Wheat 95.02 88.48 92.31 88.97 93.54 Bare soil 89.26 92.07 5.95 92.08 93.61 Beet 88.08 81.51 1.12 82.13 80.51 Rapeseed 92.79 92.07 14.37 92.28 85.13 Pea 66.91 72.32 1.86 73.13 76.83 Grass 85.46 73.71 1.42 73.71 75.36 Water 98.55 93.87 48.20 95.52 99.46 OA 91.01 87.44 34.70 87.74 89.11
Table 3. Classification accuracy of AIRSAR Flevoland 2%
View tableTable 3. Classification accuracy of AIRSAR Flevoland 2%
Category SLDA PCA LPP KNN LDA Wheat 97.93 97.26 95.50 97.35 97.44 Rapeseed 98.81 98.44 52.58 98.51 99.52 Beet 87.74 83.38 19.35 83.68 73.61 Onion 53.64 50.79 0.93 51.29 33.75 Corn 47.38 44.12 4.78 44.06 43.76 Lucerne 97.74 96.57 5.54 96.60 91.34 Barley 92.57 89.47 36.24 89.44 73.62 Flax 97.32 96.41 1.78 96.51 92.26 Fruit 96.43 92.20 10.45 92.17 93.03 Grass 70.16 65.21 15.22 65.30 53.29 Pea 68.68 73.10 0 73.01 53.74 Steam bean 41.04 39.92 14.50 39.56 20.75 Potato 97.94 97.92 36.35 98.03 97.59 OA 92.53 90.81 47.84 90.89 85.59
Tools
Get Citation
Copy Citation Text
Xinfang Xie, Xin Xu, Hao Dong, Han Wu, Luoru Li. A Semi-Supervised Dimension Reduction Method for Polarimetric SAR Image Classification[J]. Acta Optica Sinica, 2018, 38(4): 0428001
Paper Information
Category: Remote Sensing and Sensors
Received: Oct. 19, 2017
Accepted: --
Published Online: Jul. 10, 2018
The Author Email: Xu Xin (xinxu@whu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20