Forest Type Divide And Appraise Studies On The Basis Of Artificial Neural Network

The forest sets up a basic job that classification and appraisal of the ground are production of forestry, it is the foundation realizing science's fostering, dealing in and the harvest yield of forest predict. Traditional method analyze, return to with main composition analysis with quantity theory come, realize, set up to forest type go on, divide and appraise progressively. Which kind of method but no matter which kind of theory is adopted, from start to finish, set up forest ground nonlinearity and complexity relation of every factor of system regard as the simple linear relation, reflect by simplifying, reliance and indirect mathematics formula that are set up. Model set up make modeling person seem unable to do what one wishes, classify and think the result unsatisfactory often, have not found an effective solution fundamentally.In order to probe into a set of new classifying and appraising modeling technology setting up, this thesis regards stepping on a forest zone of Shaolin Temple of forest farm and large mountain range forest zone of forest farm of thin mountain as the research object, check the materials with every class II forest of primary class in the forest zone, under MATLAB systematic environment, employ LVQ nerve network and BP nerve network modeling technology, classify and appraise and study to the primary class in the forest zone standing. Set up type divide result indicate: The LVQ neural network has sentencing rate in the forest land vertical type division far to be lower than by mistake gathers the kind of analysis method, the classified result is credible and reliable.Set up type appraise result indicate: (1) Succeed in constructing 1 with BP nerve network modeling technology: S: Single shape position index model of 1, with thin mountain forest farm old mountain range to build model train and examine masson pine artificial forest class II census data (age and advantage high) of forest zone, get...