Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Simultaneously prediction of multiple wheat leaf phenotypes using hyperspectral imaging with multi-task machine and

Xiaoyu Fu¹, Lili Yao¹, Xincheng Zhang²

¹School of Information Engineering, Huzhou Normal University, Huzhou 313000, China.

Spectrochimica Acta. Part A, Molecular and Biomolecular Spectroscopy

|June 23, 2026

Summary

Related Concept Videos

Light Acquisition

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Vorinostat Rescues Cognitive Deficits in a Neuroinflammatory Mouse Model: A Study of Sex Differences and the Underlying TLR4/NF-κB Mechanism.

Neurochemical research·2026

Same author

High-speed and high-sensitivity multi-gas detection based on parallel heterodyne LITES sensor.

Light, science & applications·2026

Same author

GMFNet: A GADF-Mamba Fusion Network for Soybean Seed Hyperspectral Classification.

Foods (Basel, Switzerland)·2026

Same author

Modified Rice Bran Dietary Fiber-Based Pre-Emulsion as a Fat Replacer: Modulating Physicochemical and Sensory Properties of Emulsified Meat Gels.

Foods (Basel, Switzerland)·2026

Same author

Testican-1 activates Wnt/β-Catenin signaling to drive colorectal cancer progression.

Oncogenesis·2026

Same author

Self-Optimization With Oriented Facet Reconstruction for Universal Electrocatalytic C-N Coupling.

Angewandte Chemie (International ed. in English)·2026

Same journal

An upconversion FRET biosensor using ATRP-grafted polymer brushes as a signal amplifier for ultrasensitive detection of CYFRA21-1.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026

Same journal

A chemiluminescence sensor for ciprofloxacin detection based on copper ion and aptamer co-modified magnetic microspheres.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026

Same journal

Application of far-infrared spectroscopy for prediction of silicate mineral content in claystones and clay shales.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026

Same journal

A coumarin-based water-soluble fluorescent probe for tandem detection of Cu<sup>2+</sup> and glutathione with application in bioimaging and real sample analysis.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026

Same journal

Chromone-based thiosemicarbazone and semicarbazone as efficient fluorescent sensors for CrO<sub>4</sub><sup>2-</sup>/ Cr<sub>2</sub>O<sub>7</sub><sup>2-</sup> Ion detection.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026

Same journal

Rapid species authentication and protein prediction of porcini mushrooms using FTIR-2DCOS coupled with deep learning.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2026

See all related articles

This summary is machine-generated.

Hyperspectral imaging combined with multi-task learning enables simultaneous wheat variety identification and prediction of chlorophyll and moisture content. This spectroscopic strategy offers a feasible approach for integrated wheat leaf phenotyping and precision crop monitoring.

Area of Science:

Agricultural Science
Spectroscopy
Machine Learning

Background:

Chlorophyll content (SPAD) and leaf moisture content (LMC) are crucial physiological traits in wheat, reflecting growth, stress, and breeding potential.
Accurate monitoring of these traits and wheat variety identification are vital for precision agriculture.

Purpose of the Study:

To develop and evaluate hyperspectral imaging-based models for simultaneous wheat variety classification and prediction of SPAD and LMC.
To compare the performance of single-task and multi-task learning models, including CNN and MLT-CNN, for integrated phenotyping.

Main Methods:

Acquired hyperspectral data from four wheat varieties.
Preprocessed spectral data and developed single-task (SVM, PLS, CNN) and multi-task CNN (MLT-CNN) models.

Keywords:

Hyperspectral imaging Leaf moisture content Multi-task learning SPAD Wheat varieties

Related Experiment Videos

Employed weighting strategies (EWS, ARWS, UWS) within MLT-CNN and used SHAP for model interpretation.

Main Results:

Best single-task CNN achieved 0.75 classification accuracy, r=0.83 for SPAD, and r=0.84 for LMC.
MLT-CNN with EWS showed balanced performance: 0.69 accuracy, r=0.84 for SPAD, and r=0.81 for LMC.
SHAP analysis identified key wavelengths and interpreted spectral contributions for shared and task-specific features.

Conclusions:

Hyperspectral imaging coupled with multi-task learning presents a viable and interpretable spectroscopic method for wheat phenotyping.
This integrated approach supports precision crop monitoring by simultaneously assessing variety and physiological traits.