Research on land use/cover change of wuhan based on object oriented Image interpretation method
Higher resolution remote sensor has already become the significant instrument to obtain the change information quickly,veraciously,and comprehensively.Two different remote sensing images are used to obtain the change information by using an object oriented analyzing method in Land Use/Cover Change.Data is gathered according to the images resolution,characteristic,size and texture property of the earth surface objects.Meanwhile,the color,shape,smoothness,and compactness of the images are reviewed to form many-adjacent-pixel objects which contain more semantic information.Combined with the knowledge of land use classification and loading classifier of discriminate class,a characteristic space is defined to establish a knowledge base.Two or three different typical features are chosen to serve as the training samples,and a data base could be completed successfully.Non-segmented or error-segmented objects are adjusting into correct ones favorably.Analyzing the data,only the areas of the lands for construction purpose and reservoir & bottomland have been increased in the whole city,while the areas of arable land,river & lake land,woodland,grassland,and not exploited or developed land have been decreased in more or less degree ,which means that a tendency to expansion has become more and more significant.
Image Interpretation Classification Image Understanding Feature Generalization Land Use Mapping
Zhu Haiyan Lin Aiwen
School of Resource and Environment Science,Wuhan University,Wuhan,Hubei,China,430079;Key Laboratory for geographical information system of Ministry of Education,Wuhan University,Wuhan,Hubei,China,430079
国际会议
第16届国际地理信息科学与技术大会(16th International Conference on GeoInformatics and the Joint Conference)
广州
英文
2008-06-28（万方平台首次上网日期，不代表论文的发表时间）