Neural Network Learning of Chemical Bond Representations in Spectral Indices and Features

TL;DR

This paper demonstrates that neural networks trained on hyperspectral data learn to identify chemical bonds and spectral features, providing interpretable insights into material composition and improving classification of vegetation and polymers.

Contribution

It shows that neural network weights can be interpreted to understand spectral features related to chemical bonds, bridging physics and machine learning in hyperspectral analysis.

Findings

Neural network weights correspond to spectral differences like NDVI.

Networks can distinguish between different polymers based on learned spectral features.

Interpretable weights reveal chemical bond information in spectral data.

Abstract

In this paper we investigate neural networks for classification in hyperspectral imaging with a focus on connecting the architecture of the network with the physics of the sensing and materials present. Spectroscopy is the process of measuring light reflected or emitted by a material as a function wavelength. Molecular bonds present in the material have vibrational frequencies which affect the amount of light measured at each wavelength. Thus the measured spectrum contains information about the particular chemical constituents and types of bonds. For example, chlorophyll reflects more light in the near-IR rage (800-900nm) than in the red (625-675nm) range, and this difference can be measured using a normalized vegetation difference index (NDVI), which is commonly used to detect vegetation presence, health, and type in imagery collected at these wavelengths. In this paper we show that the weights in a Neural Network trained on different vegetation classes learn to measure this difference in reflectance. We then show that a Neural Network trained on a more complex set of ten different polymer materials will learn spectral 'features' evident in the weights for the network, and these features can be used to reliably distinguish between the different types of polymers. Examination of the weights provides a human-interpretable understanding of the network.