LSSMA: Lightweight Spectral-Spatial Neural Architecture With Multiattention Feature Extraction for Hyperspectral Image Classification.
Shujie DingXiaoli RuanJing YangJie SunShaobo LiJianjun HuPublished in: IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. (2024)
Keyphrases
- lightweight
- neural architecture
- hyperspectral image classification
- feature extraction
- hyperspectral
- hyperspectral images
- active learning
- neural network
- hyperspectral data
- feed forward
- hyperspectral imagery
- basis functions
- feature selection
- multi layer perceptron
- remote sensing
- spatial information
- dimensionality reduction
- multilayer perceptron
- multispectral
- image processing
- communication infrastructure
- training data
- principal component analysis
- satellite images
- image data
- wireless sensor networks
- feature vectors
- information content
- activation function
- spatial correlation
- reinforcement learning
- recurrent neural networks
- development environments
- dimension reduction
- change detection