Research On The Theory Of Remote Sensing In Forest Stock Volume Estimation

Forest growing stock of forest resources is an important component of the survey, the traditional survey of forest resources is a long task, labor intensive, and require a lot of funding for the work. How to make use of 3S technology, combined with a small amount of ground sample data, the establishment of regional monitoring of forest volume estimation equations to estimate forest reserves, to minimize the workload of the ground field survey has become a hot issue. The paper with examples to TM remote sensing images, topographic maps, survey of forest resources as the basic data sources, using multiple regression methods of estimation - ridge estimation The stepwise regression, partial least-squares regression, the initial establishment of forest reserves and the gray value of remote sensing, GIS information on the fit between the model and calculation method of the estimation model derived from the fitting results to be compared of remote sensing to estimate forest volume process.The main contents of this paper can be summarized as follows:The first part introduced the background of the research, significance and research at home and abroad.The second part introduced the forest stock volume estimation of the theory of remote sensing, including remote sensing estimate of forest stock volume of the basic idea, the realization of these steps; how to select the best estimation of the impact of volume of remote sensing and GIS factor factor; how to extract the establishment of reserves Estimation of the sample.The third part described the meaning of the multiple correlation which appears in forest stock volume estimates between remote sensing and GIS factors and its estimation in the forest reserves of the hazards, how to diagnose their multiple correlation, and elaborated on the relevance of dealing with multiple methods: ridge estimation, stepwise regression, partial least-squares regression.The fourth part, Application: on the basis of Remote Sensing in forest stock volume estimation theory, analyzed the Simao region of Yunnan, 7500 within the scope of the survey for a class of 129 samples, including the impact of volume estimation of the optimal factor of 18 factor choice of optimal sample extraction, fitting a variety of methods to achieve the reunification, as well as comparison of various methods of fitting.