Harnessing Spectral Libraries From AVIRIS-NG Data for Precise PFT Classification: A Deep Learning Approach.

Spectral libraries derived from AVIRIS-NG data enable 94% accuracy in plant functional type classification in Indian wildlife sanctuary.

• Why it matters: Accurate differentiation of plant species and PFTs is essential for ecological monitoring and conservation efforts, yet traditional methods lack spectral detail.
• What they did: Researchers developed a spectral library from 130 plant species using hyperspectral imaging and key spectral features, then applied machine learning classifiers to categorize five PFTs.
• The result: The Gradient Boosted Machine classifier achieved a 0.94 overall accuracy and a 0.93 Kappa coefficient, demonstrating the potential of hyperspectral data and deep learning for precise ecological classification.