Hyperspectral Image Classification with Multi-Attention Transformer and Adaptive Superpixel Segmentation-Based Active Learning

Computer Science

• Active Learning 50%
• Based Active Learning 100%
• Classification Framework 25%
• Classification Performance 25%
• Convolutional Neural Network 50%
• Deep Learning 50%
• Hyperspectral Image 100%
• Image Classification 100%
• local feature 25%
• Local Information 25%
• Receptive Field 25%
• segmentation algorithm 25%
• Small Sample Size 25%
• Spatial Relationship 25%
• Superpixel Segmentation 100%

Engineering

• Classification Performance 33%
• Convolutional Neural Network 66%
• Deep Learning 66%
• Hyperspectral Image 100%
• Image Classification 100%
• Level Feature 33%
• Local Information 33%
• Receptive Field 33%
• Similarities 33%
• Spatial Feature 33%
• State-of-the-Art Method 33%

Earth and Planetary Sciences

• Hyperspectral Image Classification 100%
• Self-Attention 33%
• State of the Art 33%