T. Jaaskelainen, R. Silvennoinen, J. Hiltunen, and J. P. S. Parkkinen, "Classification of the reflectance spectra of pine, spruce, and birch," Appl. Opt. 33, 2356-2362 (1994)
Statistical pattern-recognition methods are applied to the classification of the reflectance spectra of growing trees (Scots pine, Norway spruce, and birch). We show by using large training sets that it is possible to develop classification filters that are able to discriminate the tree types with a very high probability. Our approach may offer a reference coordinate system for multispectral remote sensing of different levels of forest damage.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Maximum, Minimum, and Average Percent Errora of Reconstructions for 1056 Spectral Reflectances
Number of Eigenvectors
Minimum Error (%)
Maximum Error (%)
Average Error (%)
Pine
Spruce
Birch
Pine
Spruce
Birch
Pine
Spruce
Birch
1
1.13
1.70
0.70
51.86
44.75
25.71
7.37
8.36
4.68
2
0.79
0.63
0.49
12.12
15.68
11.18
2.75
3.42
1.88
3
0.58
0.56
0.37
7.02
7.24
5.65
1.75
2.32
1.19
4
0.38
0.51
0.36
4.34
6.34
3.72
1.28
1.62
0.97
5
0.32
0.38
0.29
3.84
6.34
3.31
1.08
1.25
0.79
6
0.25
0.27
0.25
2.45
4.01
2.24
0.91
1.07
0.68
7
0.22
0.26
0.23
1.80
2.52
2.69
0.81
0.93
0.64
8
0.21
0.24
0.20
1.63
2.21
1.49
0.75
0.79
0.59
The errors are defined as the absolute values of the difference between the original and the reconstructed spectral reflectances averaged over the wavelength band.
Table 2
Classification of the Training Data in the Two-Class Problem with the CLAFIC Algorithm
Maximum, Minimum, and Average Percent Errora of Reconstructions for 1056 Spectral Reflectances
Number of Eigenvectors
Minimum Error (%)
Maximum Error (%)
Average Error (%)
Pine
Spruce
Birch
Pine
Spruce
Birch
Pine
Spruce
Birch
1
1.13
1.70
0.70
51.86
44.75
25.71
7.37
8.36
4.68
2
0.79
0.63
0.49
12.12
15.68
11.18
2.75
3.42
1.88
3
0.58
0.56
0.37
7.02
7.24
5.65
1.75
2.32
1.19
4
0.38
0.51
0.36
4.34
6.34
3.72
1.28
1.62
0.97
5
0.32
0.38
0.29
3.84
6.34
3.31
1.08
1.25
0.79
6
0.25
0.27
0.25
2.45
4.01
2.24
0.91
1.07
0.68
7
0.22
0.26
0.23
1.80
2.52
2.69
0.81
0.93
0.64
8
0.21
0.24
0.20
1.63
2.21
1.49
0.75
0.79
0.59
The errors are defined as the absolute values of the difference between the original and the reconstructed spectral reflectances averaged over the wavelength band.
Table 2
Classification of the Training Data in the Two-Class Problem with the CLAFIC Algorithm