Toward Generalized Change Detection on Planetary Surfaces with Convolutional Autoencoders and Transfer Learning

Keyphrases

• Change Detection 100%
• Transfer Learning 100%
• Planetary Surfaces 100%
• Convolutional Autoencoder 100%
• Autoencoder Learning 100%
• Image Quality 66%
• Surface Properties 33%
• Deep Learning 33%
• Remote Sensing Image 33%
• Large Volume 33%
• Image Sensor 33%
• Meaningful Change 33%
• Latent Representation 33%
• Mars 33%
• Surface Features 33%
• Scale Sensor 33%
• Co-registration 33%
• Illumination Image 33%
• Surface Changes 33%
• Feature Engineering 33%
• Scientific Hypothesis 33%
• Nonlinear Dimensionality Reduction 33%
• Sensor Properties 33%
• Patch Level 33%
• Change Detection Studies 33%
• Remote Sensing Dataset 33%
• Manual Features 33%
• Impact Crater 33%
• Planetary Bodies 33%
• Exploration Missions 33%
• Planetary Exploration 33%
• Illumination Quality 33%
• Major Challenges 33%
• Image Dataset 33%
• Level Changes 33%
• Detecting Changes 33%
• General Representations 33%
• Small Training Dataset 33%
• Image Differencing 33%
• Binary Code Patching 33%

Computer Science

• Transfer Learning 100%
• Autoencoder 100%
• Image Quality 66%
• Learning Approach 33%
• Dimensionality Reduction 33%
• Training Dataset 33%
• Illumination Image 33%
• Imaging Sensor 33%
• Scientific Hypothesis 33%
• Deep Learning Method 33%
• Feature Engineering 33%

Earth and Planetary Sciences

• Change Detection 100%
• Planetary Surface 100%
• Transfer Learning 100%
• Surface Property 25%
• Space Exploration 25%
• Remote Sensing 25%
• Remote Sensing Image 25%