Title :
An automatic recognition system for soil erosion based on knowledge and support vector machine
Author :
Huang, Yujie ; Wang, Guangqian ; Su, Lihong ; Liu, Zhigang
Author_Institution :
Tsinghua Univ., Beijing, China
Abstract :
Remote sensing is one of important approach of investigating soil erosion. An automatic recognition system was built for investigating soil erosion on Gushanchuan valley, which is one of tributaries of The Yellow river in China and is well known for its rugged terrain. The remote sensing recognition system enjoys two features. The one is a classification system based on expert knowledge, which recognise intensity of soil erosion by using judgment rules. The other is a support vector machine, which maps land use of the valley.
Keywords :
erosion; image classification; soil; terrain mapping; China; Gushanchuan valley; Yellow river; automatic recognition; classification system; recognise intensity; remote sensing; remote sensing recognition; rugged terrain; soil erosion; vector machine; Classification tree analysis; Expert systems; Image sampling; Land surface; Remote sensing; Rivers; Soil; Support vector machine classification; Support vector machines; Vegetation mapping;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2003. IGARSS '03. Proceedings. 2003 IEEE International
Print_ISBN :
0-7803-7929-2
DOI :
10.1109/IGARSS.2003.1294816
