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2 Indian Pines高光谱图像不同算法分类精度(1%训练样本)
|
Tab.2 Indian Pines data set classification accuracy of different algorithms (1% 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 | Alfalfa | 3 | 43 | 72.25
(5.43) | 63.26
(18.20) | 92.56
(9.21) | 87.44
(6.41) | 72.42
(30.15) | 88.35
(31.74) | 95.35
(0.00) | 95.35
(0.00) | | Corn-N | 14 | 1 414 | 47.63
(7.37) | 40.07
(5.91) | 69.99
(6.50) | 57.39
(4.59) | 61.20
(8.91) | 83.17
(8.34) | 69.12
(0.42) | 83.97
(10.41) | | Corn-M | 8 | 822 | 55.60
(15.51) | 31.33
(6.92) | 55.28
(8.60) | 63.64
(7.29) | 73.96
(18.37) | 71.08
(10.79) | 44.09
(0.11) | 74.67
(14.06) | | Corn | 3 | 234 | 35.98
(11.21) | 25.37
(8.55) | 45.04
(15.61) | 29.79
(9.89) | 59.62
(33.29) | 70.28
(12.62) | 33.42
(0.19) | 92.22
(8.31) | | Grass-M | 6 | 477 | 74.25
(12.10) | 63.59
(10.07) | 70.23
(26.81) | 77.04
(9.38) | 90.54
(12.56) | 85.61
(9.82) | 69.18
(0.00) | 89.48
(3.16) | | Grass-T | 7 | 723 | 76.13
(6.57) | 77.25
(8.63) | 85.48
(3.41) | 86.07
(13.92) | 74.85
(6.53) | 90.66
(5.52) | 98.76
(0.00) | 97.10
(1.64) | | Grass-P | 3 | 25 | 30.57
(15.22) | 89.04
(5.76) | 92.00
(12.33) | 90.80
(9.25) | 81.41
(30.42) | 52.28
(25.97) | 96.00
(0.00) | 100.00
(0.00) | | Hay-W | 5 | 473 | 93.09
(5.37) | 73.82
(12.02) | 84.90
(8.79) | 94.82
(3.02) | 98.42
(2.99) | 100.00
(0.00) | 99.79
(0.00) | 100.00
(0.00) | | Oats | 3 | 17 | 18.99
(9.95) | 68.35
(20.16) | 94.12
(10.19) | 96.47
(7.44) | 48.01
(37.62) | 27.79
(20.58) | 100.00
(0.00) | 96.47
(7.89) | | Soybean-N | 10 | 962 | 53.88
(7.46) | 49.04
(8.64) | 71.10
(5.46) | 69.27
(9.05) | 70.51
(15.60) | 72.87
(10.46) | 73.80
(2.79) | 84.30
(4.51) | | Soybean-M | 24 | 2 431 | 59.35
(4.18) | 61.04
(4.71) | 82.59
(5.06) | 77.10
(6.22) | 66.08
(9.18) | 85.91
(4.45) | 79.66
(0.13) | 93.58
(4.15) | | Soybean-C | 6 | 587 | 38.95
(8.03) | 21.85
(6.44) | 48.07
(8.57) | 39.93
(7.11) | 56.11
(22.14) | 70.91
(12.35) | 55.54
(0.76) | 80.20
(9.34) | | Wheat | 2 | 203 | 84.36
(4.03) | 77.38
(11.01) | 79.61
(12.27) | 95.81
(1.89) | 96.16
(3.87) | 80.46
(12.73) | 99.31
(0.44) | 96.75
(2.38) | | Woods | 13 | 1 252 | 84.44
(2.87) | 80.95
(6.80) | 92.54
(7.27) | 87.61
(5.30) | 87.10
(5.07) | 92.88
(1.76) | 97.78
(0.04) | 92,99
(6.50) | | Buildings | 4 | 382 | 42.01
(10.72) | 19.04
(5.30) | 36.70
(7.27) | 61.52
(13.30) | 67.86
(24.93) | 80.08
(9.07) | 20.37
(0.79) | 83.66
(6.88) | | Stone | 3 | 90 | 96.89
(5.87) | 87.00
(4.84) | 96.89
(2.41) | 71.44
(25.08) | 93.81
(20.70) | 97.97
(10.17) | 74.89
(1.49) | 98.67
(0.93) | | OA | — | — | 60.60
(2.00) | 54.87
(1.63) | 74.04
(1.37) | 71.88
(2.25) | 71.00
(2.97) | 81.84
(2.88) | 86.72
(5.32) | 89.07
(1.17) | | AA | — | — | 57.46
(2.63) | 58.02
(2.14) | 74.82
(2.74) | 74.13
(3.18) | 74.88
(5.28) | 78.14
(3.00) | 84.91
(4.79) | 87.57
(1.32) | | Kappa | — | — | 54.59
(2.16) | 48.47
(1.86) | 70.12
(1.52) | 67.88
(2.52) | 66.24
(3.70) | 79.29
(3.28) | 84.77
(6.11) | 91.21
(0.84) |
|
|
|