Inventory of Wintering Geese with a Multispectral Scanner
Figures
|
Figure 1. Mean bidirectional spectral
reflectance for Ross', lesser snow, white-fronted, and cackling Canada
geese. |
![Figure 2: Line graph showing patterned fields between lines from Fig. 1 illustrating significant differences in mean bidirectional spectral reflectance.](images/fig2.gif) |
Figure 2. Significant differences (P
< 0.05) in mean bidirectional spectral reflectance among adult white
geese (adult Ross' and adult lesser snow geese combined), white-fronted,
and cackling Canada geese. |
|
Figure 3. Bidirectional spectral for
adult white geese (adult Ross' and adult lesser snow geese combined),
white-fronted geese, turbid water, bare soil, and green winter wheat. |
|
Figure 4. Manual count of the number
of geese in a flock from simultaneously acquired aerial photographs
regressed against mixture model predictions of number of geese using
local estimate of water reflectance. |
|
Figure 5. Mixture model domain for scene
composed of white geese, dark geese, and water. Mean and covariance
(1 SD ellipse) for 3 objects of the mixture model. The proportions of
a pixel covered by white geese, dark geese, and water could be estimated
using geometry by drawing a line from each vertex through the observation
to the opposite leg of the triangle. The fraction of the line between
the intersection of the lines and the opposite leg defines the proportion
of the object at the corresponding vertex. |