Related Experiment Video
Updated: Dec 8, 2025
11:37
RGB and Spectral Root Imaging for Plant Phenotyping and Physiological Research: Experimental Setup and Imaging Protocols
Published on: August 8, 2017
16.7K
Hyperspectral imaging coupled with multivariate methods for seed vitality estimation and forecast for Quercus
Lei Pang1, Jinghua Wang1, Sen Men2
1School of Technology, Beijing Forestry University, Beijing 100083, China.
Spectrochimica Acta. Part A, Molecular and Biomolecular Spectroscopy
|September 18, 2020
Summary
Hyperspectral imaging non-destructively estimates Quercus variabilis seed vitality during germination. This method accurately predicts seed viability using spectral data, offering a faster alternative to traditional methods.
Area of Science:
- Plant Science
- Biotechnology
- Spectroscopy
Background:
- Seed vitality is crucial for agriculture and forestry.
- Traditional methods for assessing seed viability are often destructive and time-consuming.
- Developing non-destructive, rapid methods for vitality assessment is highly desirable.
Purpose of the Study:
- To investigate the feasibility of using hyperspectral imaging for estimating and forecasting the vitality of Quercus variabilis seeds.
- To identify optimal spectral bands and modeling techniques for accurate vitality assessment.
Main Methods:
- Artificially accelerated aging was used to create four distinct seed vitality levels.
- Hyperspectral data were collected during the initial 10 hours of germination at hourly intervals.
- Spectral data preprocessing included Multiple Scatter Correction (MSC) and Savitzky-Golay first derivative (SG 1st).
- Feature selection methods (SPA, CARS, GA, VIP, RF) were compared.
- Vitality estimation models were built using Partial Least Square-Discriminant Analysis (PLS-DA) and K-Nearest Neighbor (KNN).
Main Results:
- The Genetic Algorithm (GA) combined with PLS-DA yielded the optimal vitality estimation model with the highest accuracy.
- Reflectance curves of different vitality levels were plotted over time based on GA-extracted characteristic bands.
- A vitality forecast model established using 0-hour data achieved high recognition rates: PLS-DA >99% and KNN >85%.
Conclusions:
- Hyperspectral imaging provides a non-destructive method for estimating Quercus variabilis seed vitality.
- Accurate vitality prediction is feasible within a shorter timeframe using this technique.
- The study demonstrates the potential of hyperspectral imaging in seed quality assessment.
