Jove
Visualize
Contact Us

Related Concept Videos

JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies
  1. Home
  3. Nondestructive Detection Of Rice Milling Quality Using Hyperspectral Imaging With Machine And Deep Learning Regression.
  1. Home
  3. Nondestructive Detection Of Rice Milling Quality Using Hyperspectral Imaging With Machine And Deep Learning Regression.

Related Experiment Video

RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
11:37

RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols

Published on: August 8, 2017

16.2K

Nondestructive Detection of Rice Milling Quality Using Hyperspectral Imaging with Machine and Deep Learning

Zhongjie Tang1, Shanlin Ma2, Hengnian Qi1

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

Foods (Basel, Switzerland)
|June 13, 2025

View abstract on PubMed

Summary
This summary is machine-generated.

Hyperspectral imaging combined with machine learning accurately predicts rice milling quality, including brown rice rate, milled rice rate, and head rice rate. This non-destructive method aids in rice breeding and quality management.

Keywords:
brown rice ratehead rice ratehyperspectral imagingmilled rice ratemulti-task learning

More Related Videos

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images
11:49

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images

Published on: February 2, 2019

9.3K
Author Spotlight: Unraveling Plant Responses to Abiotic Stresses Using the PlantScreen Robotic Platform
06:28

Author Spotlight: Unraveling Plant Responses to Abiotic Stresses Using the PlantScreen Robotic Platform

Published on: June 7, 2024

1.7K

Related Experiment Videos

RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
11:37

RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols

Published on: August 8, 2017

16.2K
Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images
11:49

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images

Published on: February 2, 2019

9.3K
Author Spotlight: Unraveling Plant Responses to Abiotic Stresses Using the PlantScreen Robotic Platform
06:28

Author Spotlight: Unraveling Plant Responses to Abiotic Stresses Using the PlantScreen Robotic Platform

Published on: June 7, 2024

1.7K

Area of Science:

  • Agricultural Science
  • Spectroscopy
  • Machine Learning

Background:

  • Rice milling quality metrics like brown rice rate (BRR), milled rice rate (MRR), and head rice rate (HRR) are crucial for economic value.
  • Accurate and simultaneous assessment of these metrics is vital for the rice industry.

Purpose of the Study:

  • To develop and compare non-destructive methods for estimating rice milling quality attributes.
  • To evaluate the effectiveness of hyperspectral imaging combined with various machine learning algorithms for predicting BRR, MRR, and HRR.

Main Methods:

  • Hyperspectral imaging was used to collect data from two rice varieties.
  • Single-task and multi-task models including Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), Convolutional Neural Networks (CNNs), and Backpropagation Neural Networks (BPNNs) were developed.
  • SHapley Additive exPlanations (SHAP) analysis was employed to interpret model predictions.

    • Main Results:

    • Multi-task models generally showed higher prediction accuracy than single-task models.
    • BPNNs achieved high accuracy for BRR and HRR (r up to 0.9) in single-task learning, while SVR excelled for MRR.
    • Multi-task BPNNs demonstrated robust performance with r values above 0.81 for all three indicators.

    Conclusions:

    • Hyperspectral imaging coupled with machine learning and deep learning offers an effective non-destructive approach for assessing rice milling quality.
    • The findings support the application of this technology in rice breeding programs and quality control during growth management.