Research On The Large Scale Land Use Remote Sensing Alteration Method Based On Object Oriented

Land resources is essential to the human, is an important means of production. Land use and its dynamic change in the sustainable development of an impact on social economy, also have indirect effects on the global environment. Rapid, objective, accurate grasp of the status of land use data, directly related to the development of national policies. Therefore, people generally prefer to monitor land change information to have an accurate understanding of the status of land, to the scientific management of land resources. The image data of remote sensing technology provides a time sequence for land use monitoring, real-time, large scale, has become an important means of monitoring of land resources. In view of the existing situation of traditional land use change survey method, taking Jinjiang city as an example, introducing the object oriented classification method, 1:5000 large scale land use vector data using a combination of high spatial resolution remote sensing image, to carry out large scale land use dynamic change method research and application. The specific research contents and conclusions are as follows:(1) the land use change of object oriented methodComprehensive utilization of high resolution remote sensing data and land use vector data, the object oriented technology support, building land use change method, including image segmentation, multi-level multi spectral data normalization, hierarchical supervised classification, change information updating process. In image segmentation, through the comparison of different segmentation scales, ultimately determine the segmentation scale is 70 can satisfy the application of 1:5000 in Jinjiang city land use; and the traditional supervised classification methods based on pixels, into the land use layer object oriented classification method for segmentation, mixed phenomenon obviously less, avoid salt algae phenomenon, the classification efficiency and classification accuracy is relatively high.(2) Land use change monitoring application in JinJiangTaking Jinjiang city as a test region, and select the land use layer status of QuickBird remote sensing images and 1:5000 large scale, Jinjiang city land use application of monitoring changes in land use dynamic change monitoring by the construction method, higher classification accuracy, greatly reducing the field survey work. Through the statistical analysis of regional land use, we can see that the 2008-2010 years, Jinjiang city urban village and industrial land, traffic land area has a higher increased, respectively from 5975.87 hectares,1093.47 hectares of up to 6060.53 hectares,1258.65 hectares, and the land of farmland, garden, woodland, grassland area are on the decline, decreased from 3493.34 ha to 3189.26 ha. During 2008-2010, Jinjiang city in the big city development and traffic construction, by the larger volume, and a large number of agricultural land and other land into urban village and industrial land, traffic land.