The Research And Application On Boundary Optimization Algorithm In Forest Resources Vector Data

With GIS, GPS and RS Applications continued development and application, remote sensing image resolution and GPS positioning accuracy have be to increase, improving the rapid development of GIS. for the current remote sensing image segmentation widely used in agriculture, Mapping, forestry, land and resources management, urban planning, geological survey and mineral military and other fields. on the vector data boundary error also put forward higher requirements, the traditional vector data processing has been far from the border to meet user demand for high precision. Remote sensing images used in the computer storage of digital grid,The segmented object boundaries can not be smooth as outlined obtained by the artificial boundary, the boundary map is particularly evident especially when seen in different resolutions, with the display resolution increases, the target boundary of the'saw tooth Effect'is more evident.With the development of Remote sensing images and topographic maps technology, the remote sensing image classification and analysis of the terrain of the hydrological is more accuracy than artificially discrimination in some conditions, because the complexity of nature can not be fully extracted by automatedion, Precision can be improved by the artificial outline based on hydrological analysis and remote sensing image classification. How to to find a balance between outline the boundaries of artificial precision and objectivity of computer processing, is the subject of this paper. The main contents of this paper are as follows:1, After the comprehensive analysis of development status on vector data compression algorithm and vector data smoothing, the advantages and disadvantages at the principle of vector data compression and method of a comprehensive exposition were analyzed.2, Conducted in-depth research in the general vector compression algorithm compression standard evaluation, comparing the three principles in evaluation of compression performance characteristic, the quality evaluation criterion of area error was expanded, evaluation the significance about plus or minus area of polygons deviation.3, the smoothing algorithm have be studied after the analyzed on the raster to vector algorithm and smoothing vector data method. then presented a simplifies algorithm by Mean Value for the boundary vector data,based on gradient vector raster data distribution. The experimental data to prove that this algorithm can achieve better border vector data in smoothing and simplification.4, Applicated research on optimal border from the forest type survey data vector data., boundary were extracted from spatial analysis of topography and forest resources in remote sensing image classification by analyzing the DEM and remote sensing image segmentation,using Arcgis software integration tool to correct the boundary after comparison hand-painted with the use of digital vector data, concluded that it don't need to remove small polygons comprise to traditional methods,and improve the classification accuracy of the area of error, to provide a new strategy and stronger operability for the demands of high accuracy forestry.In short, the paper analysis the existing vector data compression and smoothing algorithm, then expands the criteria in the area of compression standard error, this paper also proposes a meddle value smoothing algorithm based on the probability of vector data boundary,to reduce redundancy and do not expand accumulative error as criterion. Application thise algorithm in the remote sensing image segmentation of forest inventory, based on this proposed boundary of Forest Inventory small integrated strategy to correct overlay analysis, preliminary studies have shown that comprehensive correct application of strategies to improve forestry small division of accuracy is a certain value.