A review of technical factors to consider when designing neural networks for semantic segmentation of Earth Observation imagery.
Sam KhallaghiJ. Ronald EastmanLyndon D. EstesPublished in: CoRR (2023)
Keyphrases
- semantic segmentation
- earth observation
- remote sensing data
- remote sensing
- satellite data
- superpixels
- weakly supervised
- multispectral
- conditional random fields
- scene classification
- sensor data
- pattern recognition
- object categories
- pascal voc
- remote sensing images
- image data
- object classes
- bag of words
- high resolution
- land cover
- computer vision
- data center
- hyperspectral
- database
- object class
- image processing
- satellite images
- spatial information
- supervised learning
- information extraction
- image analysis
- multiscale
- training data