递归滤波与KNN的高光谱遥感图像分类方法
|
|
涂兵, 张晓飞, 张国云, 王锦萍, 周瑶
|
Hyperspectral image classification via recursive filtering and KNN
|
|
Bing TU, Xiaofei ZHANG, Guoyun ZHANG, Jinping WANG, Yao ZHOU
|
|
表3 Salinas高光谱图像不同算法分类精度(2%训练样本)
|
Tab.3 Salinas data set classification accuracy of different algorithms (2% of training samples)
|
|
| 指标 类别 | 训练样
本/个 | 测试样
本/个 | SVM | SRC | JSRC | EMP | EPF | IFRF | LMLL | RF-KNN | | CA | Weeds_1 | 40 | 1 969 | 100.00
(0.00) | 98.36
(0.65) | 100.00
(0.00) | 99.80
(0.00) | 100.00
(0.00) | 100.00
(0.00) | 100.00
(0.00) | 100.00
(0.00) | | Weeds_2 | 73 | 3 653 | 97.19
(0.53) | 98.52
(0.45) | 99.41
(0.67) | 99.56
(0.34) | 100.00
(0.00) | 99.99
(0.02) | 100.00
(0.00) | 99.86
(0.13) | | Fallow | 38 | 1 938 | 94.60
(1.45) | 96.76
(1.21) | 99.16
(0.77) | 99.54
(0.27) | 94.84
(1.61) | 99.88
(0.08) | 99.69
(0.15) | 100.00
(0.00) | | Fallow_P | 26 | 1 368 | 97.63
(1.11) | 99.26
(0.33) | 88.67
(5.45) | 98.30
(1.20) | 98.02
(0.56) | 97.84
(0.92) | 98.26
(2.88) | 97.35
(2.34) | | Fallow_S | 52 | 2 626 | 98.55
(0.54) | 94.39
(0.68) | 84.03
(2.06) | 96.77
(0.46) | 99.94
(0.05) | 99.47
(0.98) | 99.06
(0.28) | 98.64
(0.89) | | Stubble | 79 | 3 880 | 99.97
(0.05) | 99.69
(0.10) | 98.20
(1.33) | 99.60
(0.38) | 99.98
(0.02) | 100.00
(0.00) | 100.00
(0.00) | 99.76
(0.18) | | Celery | 70 | 3 509 | 99.40
(0.31) | 99.27
(0.14) | 95.10
(2.08) | 99.58
(0.09) | 99.84
(0.17) | 99.81
(0.11) | 99.94
(0.00) | 99.93
(0.05) | | Graps | 225 | 11 046 | 74.60
(1.74) | 73.62
(1.49) | 98.47
(0.23) | 96.38
(0.91) | 84.10
(4.04) | 99.64
(0.14) | 92.72
(0.06) | 99.87
(0.19) | | Soil | 124 | 6 079 | 99.62
(0.03) | 97.89
(0.93) | 99.99
(0.01) | 99.84
(0.23) | 99.18
(0.32) | 99.92
(0.12) | 100.00
(0.00) | 100.00
(0.00) | | Corn | 21 | 3 257 | 79.07
(5.21) | 78.13
(3.72) | 89.60
(3.36) | 93.38
(1.17) | 99.21
(0.83) | 99.64
(0.42) | 89.72
(1.99) | 98.66
(0.73) | | Lettuce_4wk | 21 | 1 047 | 86.93
(4.99) | 96.58
(2.71) | 88.83
(4.86) | 96.85
(1.53) | 96.97
(1.27) | 99.20
(0.30) | 95.52
(0.77) | 97.06
(2.52) | | Lettuce_5wk | 38 | 1 889 | 97.96
(0.49) | 99.72
(0.58) | 94.55
(0.99) | 99.52
(0.98) | 99.46
(0.63) | 98.82
(1.22) | 100.00
(0.00) | 99.68
(0.40) | | Lettuce_6wk | 18 | 898 | 98.47
(0.85) | 97.34
(0.43) | 83.78
(5.46) | 98.57
(1.17) | 98.58
(1.62) | 99.01
(1.22) | 97.16
(0.13) | 91.76
(8.28) | | Lettuce_7wk | 20 | 1 050 | 86.93
(4.81) | 92.69
(2.17) | 79.62
(5.57) | 96.13
(1.75) | 98.71
(0.53) | 98.16
(1.25) | 97.52
(0.04) | 98.84
(0.66) | | Vinyard_U | 140 | 7 128 | 66.51
(5.39) | 61.41
(1.96) | 97.39
(0.47) | 94.23
(1.16) | 91.82
(2.37) | 99.88
(0.16) | 68.43
(1.31) | 99.01
(0.44) | | Vinyard_T | 36 | 1 771 | 96.88
(2.12) | 95.57
(2.75) | 99.64
(0.18) | 99.31
(0.39) | 99.75
(0.44) | 99.97
(0.05) | 97.33
(0.48) | 100.00
(0.00) | | OA | — | — | 87.57
(1.08) | 86.69
(0.54) | 95.98
(0.53) | 97.53
(0.20) | 94.78
(1.23) | 99.42
(0.10) | 93.23
(0.19) | 99.46
(0.12) | | AA | — | — | 92.14
(0.85) | 92.45
(0.54) | 93.53
(1.09) | 97.96
(0.25) | 97.53
(0.32) | 99.15
(0.10) | 95.95
(0.21) | 99.29
(0.14) | | Kappa | — | — | 86.14
(1.19) | 85.17
(0.61) | 95.53
(0.59) | 97.25
(0.22) | 94.17
(1.38) | 99.62
(0.11) | 92.44
(0.21) | 98.78
(0.54) |
|
|
|