Self-supervised Learning for Hyperspectral Images of Trees
By: Moqsadur Rahman , Saurav Kumar , Santosh S. Palmate and more
Potential Business Impact:
Helps farmers understand plant health from drone pictures.
Aerial remote sensing using multispectral and RGB imagers has provided a critical impetus to precision agriculture. Analysis of the hyperspectral images with limited or no labels is challenging. This paper focuses on self-supervised learning to create neural network embeddings reflecting vegetation properties of trees from aerial hyperspectral images of crop fields. Experimental results demonstrate that a constructed tree representation, using a vegetation property-related embedding space, performs better in downstream machine learning tasks compared to the direct use of hyperspectral vegetation properties as tree representations.
Similar Papers
Prospects for Mitigating Spectral Variability in Tropical Species Classification Using Self-Supervised Learning
CV and Pattern Recognition
Identifies plants from sky pictures better.
Self-supervised and Multi-fidelity Learning for Extended Predictive Soil Spectroscopy
Machine Learning (CS)
Helps predict soil health using light.
Label-Efficient 3D Forest Mapping: Self-Supervised and Transfer Learning for Individual, Structural, and Species Analysis
CV and Pattern Recognition
Helps identify trees faster for forests.