Development of low‐cost leaf spectroscopy based on featured wavelength extraction by fix‐bandwidth Gaussian mixture model

Light spectrum analysis of leaves plays an essential role in measuring the contents of carbohydrates and other important compound like proteins. And it is also useful in evaluating the status of vegetation by remote sensing. Here, an approximation method considering the optical response characterist...

Full description

Saved in:
Bibliographic Details
Published in:IET science, measurement & technology measurement & technology, 2023-06, Vol.17 (4), p.158-166
Main Authors: Long, Xingming, Zhang, Ruoshuang, Zhou, Jing
Format: Article
Language:English
Subjects:
condition monitoring
distributed sensors
Gaussian distribution
light propagation
measurement systems
optical microscopy
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Light spectrum analysis of leaves plays an essential role in measuring the contents of carbohydrates and other important compound like proteins. And it is also useful in evaluating the status of vegetation by remote sensing. Here, an approximation method considering the optical response characteristics of multispectral silicon (Si) sensor for the reflection or absorption spectra of leaves is proposed. The light spectrum is analysed by Gaussian mixture models (GMM) with fixed bandwidth firstly, and then the optimal model parameters are derived and validated to balance the complexity of Si sensor and the 2‐norm deviation of spectra, and next a low‐cost framework for the leaf spectroscopy with wireless node is illustrated according to the specific centre‐wavelengths and bandwidths of the GMM, and finally, an experimental prototype with 18‐channel Si sensor node and the developed mobile APP is demonstrated. The simplified strategy and its realization for leaf spectroscopy cast light on large‐scale applications in future agriculture. A simplified method for measuring leaf spectrum paves the way towards its applications in precision agriculture. The spectra of leaves are approximated by the Gaussian mixture model with fixed bandwidths. A framework and its prototype comprising a wireless spectral sensor node and a mobile android app are designed and developed. Compared with the FBG sensors in the conventional instrument, the leaf absorption coefficients extracted by the low‐cost IoT‐based measurement of leaf spectrum are discussed.
ISSN:1751-8822
1751-8830
DOI:10.1049/smt2.12139