Abstract
Digital analysis method of multispectral air photos or sensor information will be recognized as the basic data handling way on quantitative processing work.
In this report, the author described the result of a test on automatic forest interpretation by multispectral air photos. The classified items were 11. The total estimation was down about 71% correctness. 94% of the hard wood forest was identified correctly. But the identification correctness of “Hinoki (Chamaecyparis obtusa) ” forest was only 51% due to their mixture with “Sugi (Cryptomaria japonica) ”. The size determination of the observation slit was the fundamental factor to be discussed in further studies.
