Open AccessAn extraction method of cultivated land in karst area based on Landsat time series
Author(s) -
Limeng Wang,
Jiasheng Wang
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/734/1/012022
Subject(s) - cultivated land , karst , geography , land use , remote sensing , distribution (mathematics) , forestry , physical geography , cartography , hydrology (agriculture) , archaeology , geology , agriculture , ecology , mathematics , mathematical analysis , geotechnical engineering , biology
It is of great significance to grasp the distribution of cultivated land in karst area for protecting cultivated land and maintaining regional food security. Based on Landsat 8 OLI remote sensing images, this paper took Fuyuan County of Yunnan Province as an example, used a classification method combining multi-feature construction with random forest to extract the distribution of cultivated land in Fuyuan County in 2019. The experimental results showed that the overall accuracy was 0.93, and the Kappa coefficient was 0.90. This method could reduce the interference of mixed pixels to a certain extent and had a better classification effect. The cultivated land area in Fuyuan County was 1200.42 km2. The cultivated land was distributed in all townships of the whole county, and it was mainly concentrated in the northwest, central and part of the southern regions of Fuyuan County, such as Housuo Town, Zhong’an Town, Dahe Town, Yingshang Town, Zhuyuan Town, and Shibalianshan Town.