Application research of technique with defining classification signatures based on virtual reality on remote sensing investigation of soil and water conversation
TUN Qing-;Wang-Dong-Mei-;Zhou-Xin-Cheng-;Qin-Zhi-Jiang-;Ding-Su-Rong
1.College of Soil and Water Conservation,Beijing Forestry University,Key Lab.of Soil & Water Conservation and Desertification Combating,Ministry of Education,100083,Beijing;2.Institute of Yellow River Water Resource Science, Yellow River Water Resource Committee,450003,Zhengzhou;3.Water Resource Bureau of Guantao County,Hebei Province, 057750,Guantao,Hebei;4.Agriculture Institute of Chifeng,Inner Mongolia,024031,Chifeng,Inner Mongolia:China
Abstract:For improving the classification accuracy on remote sensing investigation of soil and water conversation,the Virtual Reality Environment for watershed of Yiluohe River was generated based on Virtual GIS software platform.Using the satellite images of SPOT5 2.5?m,the training sites were selected and the classification signatures was defined under the 3D visible viewer.Comparing with the traditional method under 2D viewer,the results shown the total precision of the former can reach at 95.48%,and the latter only 88.19%.Our research has illuminated that the technique of Virtual Reality can enhanced the capacity of perceivability on satellite images and increase the classification accuracy.Finally,it presented that the Virtual Reality technique have the huge potentiality on satellite images classification.
吴卿1,2;王冬梅1;周心澄1;秦志强3;丁素荣4. 基于虚拟现实的分类模板定义技术在水土保持遥感调查中的应用[J]. 中国水土保持科学, 2006, 4(6): 56-60.
TUN Qing-;Wang-Dong-Mei-;Zhou-Xin-Cheng-;Qin-Zhi-Jiang-;Ding-Su-Rong. Application research of technique with defining classification signatures based on virtual reality on remote sensing investigation of soil and water conversation. SSWCC, 2006, 4(6): 56-60.