An Object-Oriented CNN Model Based on Improved Superpixel Segmentation for High-Resolution Remote Sensing Image Classification.
Zhiqing LiErzhu LiAlim SamatTianyu XuWei LiuYihu ZhuPublished in: IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. (2022)
Keyphrases
- remote sensing
- high resolution
- image classification
- remote sensed images
- fully unsupervised
- remotely sensed data
- image analysis
- medical imaging
- change detection
- hyperspectral remote sensing
- multispectral
- remote sensing images
- high spatial resolution
- image processing
- remote sensing imagery
- satellite imagery
- remote sensing data
- super resolution
- hyperspectral
- spatial resolution
- image fusion
- satellite images
- feature extraction
- segmentation algorithm
- remotely sensed images
- image segmentation
- image regions
- multiscale
- segmentation method
- land cover
- digital image analysis
- medical images
- high frequency
- remotely sensed
- hyperspectral images
- multi spectral images
- satellite data
- automatic image registration
- land cover classification
- geographical information systems
- magnetic resonance images
- image features
- high quality
- pixel classification
- earth observation
- pattern recognition
- region growing
- multiresolution
- sensor data
- machine learning