Fig 1.
Study area.
Table 1.
GF-1 satellite PMS data parameters.
Fig 2.
Research workflow.
Fig 3.
NDVI histogram in 2017.
Fig 4.
2017 Vegetation extraction results.
The base map of the LandSat8 OLI images with the following composition:R (4),G (3) and B (2). The LandSat8 OLI images were downloaded from USGS National Map Viewer.
Table 2.
Optimal thresholds in 2014–2017.
Fig 5.
Multiresolution segmentation (SS = 66, WS_1 = 0.5, WC = 0.7): (a) variance curve, (b) optimal segmentation scale and (c) image object.
Fig 6.
Multiresolution segmentation (SS = 206, WS_1 = 0.5, WC = 0.7): (a) variance curve and (b) image object.
Fig 7.
Multiresolution segmentation (SS = 66, WS_1 = 0.5, WC = 0.1): (a) variance curve and (b) image object.
Fig 8.
Multiresolution segmentation (SS = 206, WS_1 = 0.5, WC = 0.1): (a) variance curve and (b) image object.
Table 3.
J-M distances of part features (Homogeneity(Homo)).
Table 4.
Correlation coefficients and J-M distances of feature set (homo(homogeneity) dis(dissimilarity), neig(neighbors)).
Table 5.
Training and validation samples.
Fig 9.
The base map of the LandSat8 OLI images with the following composition:R (4),G (3) and B (2). The LandSat8 OLI images were downloaded from USGS National Map Viewer.
Fig 10.
Extraction results in (a)2014 (b)2015 and (c)2016.
The base map of the LandSat8 OLI images with the following composition:R (4),G (3) and B (2). The LandSat8 OLI images were downloaded from USGS National Map Viewer.
Table 6.
Accuracy assessment of 2017 classification result (OA(Overall accuracy), PA(Producer accuracy), UA(User accuracy), ML(Maximum likelihood)).
Table 7.
Accuracy assessment of 2014–2016 SVM classification result (OA(Overall accuracy), PA(Producer Accuracy), UA(User Accuracy)).
Fig 11.
Spatial distribution during (a)2014-2015 (b)2015-2016 and (c)2016-2017.
The base map of the LandSat8 OLI images with the following composition:R (4),G (3) and B (2). The LandSat8 OLI images were downloaded from USGS National Map Viewer.
Fig 12.
Spatial variation from 2014 to 2017.
Fig 13.
Area change from 2014 to 2017.
Table 8.
Tea plantation area changes during 2014–2017.
Fig 14.
Correlation coefficients of texture features.
Table 9.
Accuracy assessment of 2017 binary classification and multi-classification result (OA(Overall accuracy), PA(Producer accuracy), UA(User accuracy), ML(Maximum likelihood)).