CAJ | 학술논문

对城市用地进行分类研究，可为城市土地的合理规划布局与有效管理提供即时信息支撑。利用Quik-Bird高分辨率遥感影像，采用面向对象的影像分析技术与多尺度影像分割技术．运用eCogni—tion软件对武汉市城市中心区用地进行分类研究。针对研究区各用地类型．构建优化分类模型，选择适宜的影像分割尺度和分割特征、按照合理的顺序逐步提取城市用地分类信息。评价结果表明：该方法分类速度快、精度高，为高分辨率遥感影像城市用地分类研究提供了新思路。
대성시용지진행분류연구，가위성시토지적합리규화포국여유효관리제공즉시신식지탱。이용Quik-Bird고분변솔요감영상，채용면향대상적영상분석기술여다척도영상분할기술．운용eCogni—tion연건대무한시성시중심구용지진행분류연구。침대연구구각용지류형．구건우화분류모형，선택괄의적영상분할척도화분할특정、안조합리적순서축보제취성시용지분류신식。평개결과표명：해방법분류속도쾌、정도고，위고분변솔요감영상성시용지분류연구제공료신사로。
The classified research into urban land prov ides the rational layout and efficient management with instant information support. Using Quik-Bird high resolution remote sensing images and object-oriented and multi-scale image segmentation and adopting the eCognition software, this paper conducts classified research into the downtown area of Wuhan City. According to the different types of the land in the city, the researchers create the module of classification, select proper scale parameter and segmentation characters, and then take proper sequence to extract the urban land cover information. Accuracy assessment result indicates that this method has a precise result and short course,providing a new way for the research on the urban land cover classification.