Hyperspectral Analysis of Grain Year Identification
author: Callum
2025-01-22
1. Introduction
The identification of the storage year of grains is crucial for ensuring food quality and safety. Taking wheat as an example, the germination rate of wheat seeds is closely related to its storage year. As the storage year increases, the germination rate of wheat seeds typically decreases, which in turn affects both the yield and quality of wheat. Therefore, accurately identifying the storage year of grains is essential for agricultural production and food reserve management. Traditional methods for identifying grain storage years rely primarily on manual sensory judgments, such as observing the grain’s appearance or smelling it. However, these methods are time-consuming, labor-intensive, and prone to errors, making it difficult to meet the modern agricultural demand for fast and accurate grain year identification.
With the continuous development of spectral technologies, hyperspectral imaging has emerged as a promising solution, offering new technical means for grain year identification. Hyperspectral imaging combines the advantages of imaging and spectral technologies, allowing for the simultaneous acquisition of both image and spectral data that reflect the external characteristics, internal physical structure, and chemical composition of the sample. This technology is non-destructive, fast, and accurate, and has been widely applied in non-destructive testing of agricultural products.
This study aims to use hyperspectral imaging technology to identify the storage years of grains (such as wheat, corn, etc.). By collecting hyperspectral data from grains of different storage years, extracting their characteristic spectral information, and constructing year identification models, rapid and accurate identification of grain years can be achieved. This not only helps improve the scientific management of grain reserves but also contributes to the sustainable development of grain production.
2. Technical Approach and Main Content
2.1 Sample Selection and Data Collection
Steps:
1. Select representative grain samples from different years, such as wheat, corn, etc.
2. Use a hyperspectral imaging system to collect the hyperspectral data of the samples. It is crucial to ensure consistency in the collection conditions to avoid external factors influencing the results.
Advantages:
Non-destructive testing ensures the integrity of the samples and allows for the acquisition of spatial distribution information across the entire sample. This helps to identify any internal inhomogeneities.
Non-destructive testing ensures the integrity of the samples and allows for the acquisition of spatial distribution information across the entire sample. This helps to identify any internal inhomogeneities.
2.2 Data Preprocessing
The collected hyperspectral data undergoes preprocessing operations such as radiometric calibration and noise removal to improve data quality. The calculation formula for radiometric calibration is as follows:
Figure 1 Hyperspectral Image of Grain
Advantages:
This step allows for the acquisition of detailed spectral information, containing rich chemical and physical data. It enables the distinction of subtle differences in components and the detection of complex compositions.
By using methods like Principal Component Analysis (PCA), redundant information between bands is removed, and multi-band image data is compressed into a smaller number of effective transformed bands.
Figure 2 PCA-Based Synthetic Color Image of Grain Seeds from Different Years
2.3 Feature Extraction and Selection
Using PCA transformation for RGB color synthesis, we can more objectively observe the differences in the images of wheat seeds from different years. To further analyze these differences, we extract the features of wheat seeds from various years, such as the mean and standard deviation. The variations in the mean and standard deviation of wheat seeds from different years are shown in the figure below.
From the figure, it is evident that the spectral curves of wheat seeds from different years are very consistent, both in terms of the mean and the standard deviation. Looking at the mean, the wheat seeds from different years show relatively distinct differentiation around 500 nm. For the standard deviation, differentiation is more evident at 700 nm. Therefore, in this study, these two bands are used to construct NDVI and EVI spectral indices, and the formulas for calculating NDVI and EVI are as follows:
2.4 Results Analysis
Taking the NDVI spectral index as an example, we analyze the image display differences of wheat seeds from different years. In the left image (excluding 2011) and the right image (excluding 2012), the image colors of wheat seeds from other years show a regular pattern of change. In the NDVI image, wheat seeds from earlier years appear blue, those from intermediate years appear green, and the most recent seeds appear red.
For modeling data, wheat seeds from the years 1995, 2008, 2009, 2010, 2012, 2013, and 2014 (7 years in total) were used, while wheat seeds from 2010, 2011, 2013, and 2014 (4 years) were used for validation. Using the NDVI and EVI spectral indices, we built a model to predict the year of wheat seeds and verified the reliability of the model.
2.5 Conclusion
The raw imaging data of wheat seeds collected by the hyperspectral sorter were preprocessed, including radiometric calibration and noise removal. The MNF method was used for noise removal. To compress the multi-band image data into a more effective set of fewer transformed bands, Principal Component Analysis (PCA) was applied to the preprocessed hyperspectral data. Based on the analysis of the mean and standard deviation, we selected the visible spectrum bands at 500 nm and 700 nm to construct the NDVI and EVI spectral indices. As the storage year of wheat seeds increased, the NDVI and EVI values also increased. The coefficient of determination (R²) for NDVI with respect to the corresponding year was 0.7624, while for EVI it was 0.8585, indicating that the model built using EVI had a higher R² than that of NDVI. The NDVI modeling equation is y=−118.48x+2051.5, and the EVI modeling equation is y=−9.312x+2027.7. Using these models, to predicted the year of wheat seeds for 2010, 2011, 2013, and 2014. Both NDVI and EVI models gave the exact predicted years matching the actual years.
UAV-Based Hyperspectral Imaging for Wheat Stripe Rust Monitoring
Development and Prospects of Hyperspectral LiDAR for Earth Observation
Related Article
Discover how UAV‑based hyperspectral imaging and deep learning can monitor SDI, TOC, and TEP – key membrane fouling indicators for seawater desalination during harmful algal blooms. Based on a recent Water Research study, this article explores how Optosky's ATH9010 enables spatial risk mapping and proactive intake management.
Applications | ATH9010 Enables Water Quality and Membrane Fouling Monitoring for Desalination During Algal Blooms
Choosing between 785 nm and 1064 nm for Raman? This guide explains the physics of fluorescence, compares signal strength and fluorescence suppression, and provides a step‑by‑step decision tree. Real‑world drug detection and pesticide examples show why wavelength matters – and how dual‑wavelength coverage offers the ultimate solution.
785 nm or 1064 nm? Fluorescence Is the Dividing Line
Inspired by the 2026 Science paper from Zhang Jun’s team on a video‑rate on‑chip hyperspectral microsystem, Optosky’s ATH series moves beyond lab‑only tools. We deliver tailored, real‑world hyperspectral systems that integrate compact hardware, edge AI, and application‑specific algorithms—turning raw spectral data into actionable decisions, right where you need them.
From Science to the Field: Custom Hyperspectral Solutions by Optosky
See how 40 Optosky NY3300Pro multispectral units were deployed across a large‑scale demonstration farmland in Northwest China. With 9‑band imaging, solar power, 4G transmission, and automated data analytics, this solution enables precise growth monitoring, early pest detection, and water‑fertiliser optimisation – a true leap toward smart agriculture.
Case Study | Bulk NY3300Pro Installation Transforms Smart Farming