International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
Red: Band 110 (850nm).
Blue: Band 35 (550nm)
Figure 3. The RDACS dataset , September 7, 1999
including the Pixel Purity Index (PPD, an n-Dimensional
Visualizer, Spectral Angle Mapper (SAM) and Binary
Encoding were utilized to map the study datasets.
For detailed classification, the n-Dimensional Probability
Functions (nPDF) approach was used (Cetin, 1990; Cetin
and Levandowski, 1991; Cetin ef al, 1993). The nPDF
techniques is an interactive
Figure 4. a) Gray-scale placards, b) RDACS imafe showing
the location of the placards, c) GPS measurement, d)
sycamore leaves; healthy and under stress, and e) Full range
(350-2500nm) field spectroradiometer used in this study
image analysis technique, which overcomes many of the
inherent limitations of traditional classifiers. The techniques
has applications in three broad areas: data visualization,
enhancement and classification. For data visualization, nPDF
1309
LBL Overstory Vegetation Spectra
Reflectance (%)
(nm)
Figure 5. Spectra of the vegetation species in LBL
provides a method for transforming multiple bands of data in
a predictable, and scene-independent way. These
transformations may be designed so as to enhance a particular
cover-type, or to give the best visual representation of the
multi-band image data. Spectral frequency plots of the nPDF
components give a spectral view of data distribution that can
be used to investigate the number and distribution of spectral
classes in a high dimensional data set. In addition, these plots
are used n a non-parametric classification of the image for
discrimination of discrete classes, as well as for classes that
are mixtures at the sub-pixel scale. In a mixed deciduous and
coniferous forest an nPDF Deciduous Forest Index showed a
high correlation with percent deciduous vegetation determined
from field surveys.
The nPDF approach may be explained using a cube model. A
generalized distribution of highly correlated digital remotely
sensed data in three dimensional feature space is shown in
Figure 1. In three-dimensional feature space the feature
vector is defined by X=[ x1,x2,x3]. The location of a point
within the range of the total possible measurement space can
be described by the distances to the two corners of the cube
shown in Figure 6. They are:
2 2 2.12
Di = (x, + X x y (1)
Zu 2 2 A2
D5- [x *XQ,t(R-x) ] (2)
For the multi-dimensional case, the feature vector is defined
by X=[x1,x2.x3.....x,], where n is the dimension of the data
and R is the maximum possible range of the data (255 for 8
bit data.) When a hyper-dimensional cube is used, the vector
magnitudes (the distances to the two corners) for n-
dimensional data are: