Forest/non-forest mapping using ENVISAT ASAR data in Northeast China

Author

Huang, Yanping ; Ling, Feilong ; Wu, Bo ; Bai, Lina ; Tian, Xin

Author_Institution

Key Lab. of Spatial Data Min. & Inf. Sharing of Minist. of Educ., Fuzhou Univ., Fuzhou, China

fYear

2011

fDate

June 29 2011-July 1 2011

Firstpage

381

Lastpage

385

Abstract

Large scale forest mapping and change detection plays a significant role in the study of global change, particularly in the research of carbon source and sink. This paper presents results from forest/non-forest classification using ENVISAT-ASAR data. Both pixel-based and object-based classification method were developed for ASAR HH/HV images acquired on a single date. For the object-based classification, two different strategies were proposed: rule-set and threshold-ratio. Using as reference a land use map derived from Landsat TM images acquired in 2000, the accuracy of the forest/non-forest map from ASAR AP data has been found to meet the requirements of mapping the Northeast Chinese forests at large scale.

Keywords

geophysical image processing; image classification; remote sensing by radar; vegetation mapping; AD 2000; ENVISAT ASAR data; HH/HV images; Landsat TM images; Northeast China; carbon sink; carbon source; change detection; global change; nonforest mapping; object-based classification; pixel-based classification; Accuracy; Feature extraction; Pixel; Robustness; Satellites; Scattering; Urban areas; SAR; classification; forest; object-based; segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Spatial Data Mining and Geographical Knowledge Services (ICSDM), 2011 IEEE International Conference on

Conference_Location

Fuzhou

Print_ISBN

978-1-4244-8352-5

Type

conf

DOI

10.1109/ICSDM.2011.5969069

Filename

5969069