Study On The Forest Resources Volume Monitor Of Saihanba Mechanical Forest Farm Based On GPS, GIS And RS Technology

Based on the Remote Sensing (RS), Geographical Information System (GIS) and Global Position System (GPS), take 10m resolution Remote Sensing Images of SPOT5 as the basic data resources and combine ground quadrate investigation, ground location data analysis and other documents, use relative math theories and methods, such as mean residual, ridge regression, stepwise regression et al. to discuss the ways of SPOTS Remote Sensing data pre-processing, forest classification with Remote Sensing , forest volume estimation based on Remote Sensing in order to provide the scientific basis and the technical method. The main research content and achievement have been obtained as follows:(1)Because of the characteristics of Saihanba mechanical forest farm,forest resources monitoring makes use of the method of SPOT5 Remote Sensing data pre-processing. The technical processes in the actual application as follows: the optimal band combination(2-4-1), Image Enhancement(Histogram Equalization,Principal Component Analysis Transformation,Tasseled Cap Transform,Geometric Rectification), orthophoto correction.(2) The classification methods of Remote Sensing Images are analyzed as follows:â‘ discuss the artificial interpretation classification ways particularly ;â‘¡in order to overcom the phenomenon of different spectra characteristics with the same object we division the trees for many subclasses according to the site conditions and tree age and only the training sample enough can the precision of supervised classification is high;â‘¢supervised classification methods are used for Remote Sensing Images of the experimental area and utilize three supervised classification methods to classify the Remote Sensing Images. By way of the Error Statistical Analysis of Matrix, the result of Maximum Likelihood is the most suitable tool to Saihanba mechanical forest farm.(3) Factors which affect the estimation of the stand volume. Many experts have done so lots of studies, but the estimation models exist so many differences due to the differences among data resources and research areas. So applicability need to be further studied and discussed. In this paper, the estimation factors which affect the stand volume are classified to two kinds. One is the different plots' gray value and band ratio from Remote Sensing Images; another is site condition features of stands. Estimation model establishment method of the forest volume was inquired deeply and in detail and finished test and correction of a series of models. On the basis of data analysis, the relative Remote Sensing factors and qualitative factors were selected to establish the optimal multivariate linear regression models of the forest volume. The results show that the estimation models of different tree species have significant linear relationship and higher correlation coefficients (>75%). This can meet the accuracy requirements of the forestry production.