Canopy characteristic scale is a basic concept of quantitative remote sensing of vegetation
but also a very important feature
expression is of great significance for his physical definition and quantitative. The canopy characteristic scale physical c onnotation understanding and modeling mathematical expression
is the basic of research object linear and non-linear mixed
and it is the premise of optimal scale of observation objects. From the ray radiation transport point of view
there is a characteristic scale nonlinear mixed into the linear mixed transition
incident radiation between the transition characteristic scales of the object is independent of the optical properties
which can better describe the canopy group
choose the appropriate scale can get twice the result with half the effort to make remote sensing data.Base on the physical definition of optical radiation transmission proposed canopy characteristic scale
we established a mathematical calculating model for the canopy characteristic scale
and introducted the inverted geostatistics index model
put forward the calculation method of canopy characteristic scale based on analysis of local variance. Using the forest area of high resolution i mage
the canopy characteristic scale model proposed in this paper provide a quantitative validation. We chose 19 areas of higher density and vigorous growth forest as a study area
all the study areas with the same planting and regular distributed
and approximately at the same stage of the operation. Data come from the Google Earth aerial photography data
with a resolution of 0. 3 m
the data from two regions in Macon and Locke.The canopy characteristic scale model for analysis of data from the two regions
There is a certain correlation between the canopy characteristic scale model calculated and forest spacing measured
the linear correlation coefficient of 0. 95
that showed spacing and vegetation canopy characteristic scale of the two regions is the presence of certain stable relationship
canopy characteristic scale calculation of canopy characteristic scale model was about 1. 25 times the spacing.Canopy features scale models thesis
which can be in remote sensing images of forest vegetation growing season to choose their appropriate characteristic scale
canopy characteristic scale model is reasonable and feasible. Firstly
mixed pixel above the canopy characteristic scale belong to linear mixed
for its various transformations
you can use linear mixed pixel method to solve various scale difference or scale effect problems
Secondly
canopy characteristic scale is a basic object scale
when vegetation p arameter population sampling（ such as leaf area index）
canopy spectrum measurement
the sampling plots close to or greater than the canopy characteristic scale
can reflect the vegetation canopy component information; When the resolution is close to or below the canopy characteristic scale
ground surrounding pixel block or cross radiation effects to a minimum
arrive downward radiation level surface solar radiation spectrum and the corresponding pixel under the surface of the ground
ground reflectance can be calculated accurately.