Abstract:

In order to defining the remote sensing factors influencing forest canopy density effectively,based on the research of predecessors,the magnitude of information on each wave bands and ratio wave bands in common use of TM remote sensing image,the multi-collinearity existed between each other and its unfavorable influence to the estimation of canopy density were studied and analyzed in this paper. To certain monitoring area and the remote sensing ratio wave bands defined,by means of the criterion of mean residual sum of squares to carry on filtering of remote sensing factors,finally the important remote sensing wave bands influencing canopy density estimation are confirmed,meanwhile the operable software system is programmed.

Key words: remote sensing ratio wave band, variance expands factor, magnitude of information, multi-collinearity