Deep-learning-based latent space encoding for spectral unmixing of geological materials

Keyphrases

• Deep Learning 100%
• Unmixing 100%
• Spectral Unmixing 100%
• Geological Materials 100%
• Latent Space 100%
• Space Encoding 100%
• Denoising 60%
• Encoder Networks 60%
• Hyperspectral 40%
• Joint Optimization 40%
• Proposed Strategy 20%
• Spectral Response 20%
• Spatial Resolution 20%
• Model Sensitivity 20%
• Trainable 20%
• Endmember 20%
• Evaluation Measures 20%
• Physical Significance 20%
• Network Parameters 20%
• Airborne Hyperspectral Images 20%
• Interpolation Method 20%
• Label Information 20%
• Simulated Dataset 20%
• Mixed pixel 20%
• Convolutional 20%
• Data Imputation 20%
• Cuprite 20%
• Number of Training Samples 20%
• Dynamic Time Wrapping 20%
• Feature Characterization 20%
• Encoding Model 20%
• Interpolation Kernel 20%
• Cross-entropy Loss 20%
• Unmixing Model 20%
• Spectral Denoising 20%
• Generative Encoding 20%
• Denoising Data 20%
• Instantaneous Field of View 20%

Engineering

• Joints (Structural Components) 100%
• Deep Learning 100%
• Limited Number 50%
• Spectral Response 50%
• Spatial Resolution 50%
• Physical Significance 50%
• Good Result 50%
• Endmembers 50%
• Hyperspectral Image 50%
• Network Parameter 50%
• Cuprite 50%
• Instantaneous Field of View 50%

Earth and Planetary Sciences

• Spectral Unmixing 100%
• Coder 75%
• Interpolation 50%
• Spatial Resolution 25%
• Field of View 25%
• Hyperspectral Image 25%