ECGAN: An Improved Conditional Generative Adversarial Network With Edge Detection to Augment Limited Training Data for the Classification of Remote Sensing Images With High Spatial Resolution.
Baikai SuiTao JiangZhen ZhangXinliang PanPublished in: IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. (2021)
Keyphrases
- remote sensing images
- high spatial resolution
- remote sensing
- multispectral
- land cover classification
- landsat etm
- multi spectral images
- change detection
- training data
- edge detection
- supervised learning
- multispectral images
- spatial resolution
- high resolution
- training set
- spatial and temporal resolution
- satellite images
- remote sensing data
- urban areas
- video camera
- pattern recognition
- feature extraction
- discriminative methods
- low spatial resolution
- hyperspectral data
- image fusion
- hyperspectral
- multiscale
- satellite imagery
- feature space
- unsupervised learning
- remote sensing imagery
- feature vectors
- multiresolution
- high quality
- machine learning
- semi supervised
- image analysis
- feature selection