Optimized Sample Selection in SVM Classification by Combining with DMSP-OLS, Landsat NDVI and GlobeLand30 Products for Extracting Urban Built-Up Areas.
Xiaolong MaXiaohua TongSicong LiuXin LuoHuan XieChengming LiPublished in: Remote. Sens. (2017)
Keyphrases
- svm classification
- sample selection
- urban areas
- support vector machine
- multispectral
- remote sensing images
- support vector machine svm
- satellite images
- remote sensing
- support vector
- kernel function
- classification accuracy
- image classification
- remote sensing data
- change detection
- svm classifier
- active learning
- classification method
- land cover
- fuzzy clustering
- neural network
- multiclass classification
- supervised learning
- small number
- feature space
- training data
- feature selection
- machine learning