Effect of Training Class Label Noise on Classification Performances for Land Cover Mapping with Satellite Image Time Series.
Charlotte PelletierSilvia ValeroJordi IngladaNicolas ChampionClaire Marais-SicreGérard DedieuPublished in: Remote. Sens. (2017)
Keyphrases
- land cover
- classification performances
- satellite images
- remote sensing
- remote sensing data
- multispectral
- feature selection
- change detection
- class label noise
- remotely sensed
- remote sensing images
- satellite data
- landsat tm
- remotely sensed images
- satellite imagery
- high resolution
- remotely sensed imagery
- geographic information systems
- digital elevation models
- supervised classification
- urban areas
- hyperspectral
- high spatial resolution
- classifier ensemble
- machine learning
- supervised learning
- training set
- support vector
- decision trees
- computer vision