This innovative device addresses key limitations of traditional multispectral sensors, which often suffer from complex optics and slow data processing. By incorporating mosaic filters and a spectral crosstalk correction method, SMICGS enhances the accuracy of crop monitoring while offering high-throughput capabilities for rapid, non-destructive assessments.
Monitoring crop growth is vital for securing yield and quality, yet traditional approaches rely on destructive sampling or subjective field surveys, both of which are time-consuming and often inaccurate. Spectral imaging has emerged as a transformative alternative, allowing researchers to link crop spectral signatures with growth characteristics for non-destructive, real-time estimation. Despite progress, current multispectral sensors frequently depend on multi-lens systems and offline data processing, creating bottlenecks in efficiency and scalability. These challenges limit practical applications for dynamic farm environments where immediate crop-growth information is critical. Addressing these technological gaps, the new SMICGS sensor integrates optical simplification with embedded real-time processing, advancing the future of UAV-based crop monitoring.
A study (DOI: 10.1016/j.plaphe.2025.100056) published in Plant Phenomics on 20 May 2025 by Dong Zhou & Yan Zhu’s team, Nanjing Agricultural University, can accurately predict critical crop growth indicators, marking a significant step forward in precision agriculture.
The study introduces a new UAV-based snapshot multispectral imaging crop-growth sensor (SMICGS) designed to enhance crop monitoring efficiency. The sensor utilizes mosaic filters tuned to crop-sensitive spectral bands, enabling multiband co-optical imaging without the need for complex multi-lens assemblies. The spectral calibration process revealed minimal deviation between preset and measured wavelengths, with a maximum difference of only 0.43 nm, demonstrating the sensor’s precision. To address spectral crosstalk, the team developed a correction method that effectively reduced interference between adjacent channels, improving spectral accuracy. After correction, reflectance errors were significantly reduced, from 26.49% to 6.47%, confirming the method's effectiveness. Additionally, the study integrated machine learning, specifically random forest (RF) algorithms, to model growth indicators such as leaf area index (LAI) and above-ground biomass (AGB) in wheat and rice using SMICGS data. The models showed high predictive accuracy, with R² values of 0.93 for rice AGB and 0.89 for LAI, and 0.85 and 0.81 for wheat, respectively. Root mean square error (RMSE) values for both crops were low, further validating the sensor’s capability to predict critical crop growth parameters. The system’s signal-to-noise ratio remained above 100 dB, confirming its robustness. When compared to commercial sensors like RedEdge, SMICGS demonstrated reliable predictions, offering a streamlined structure and integrated real-time data interpretation. These features highlight SMICGS's potential for precise, non-destructive, and scalable crop monitoring, making it an invaluable tool for precision agriculture.
The SMICGS system represents a powerful advancement for precision agriculture. By combining UAV mobility with streamlined multispectral imaging and embedded data processing, it enables farmers, researchers, and policymakers to obtain real-time insights into crop growth, nutrient status, and biomass accumulation. This capability reduces reliance on destructive sampling, lowers labor costs, and accelerates decision-making for irrigation, fertilization, and pest management. In regions facing resource constraints or climate-induced risks, the device provides a reliable foundation for boosting productivity while promoting sustainable practices. Beyond wheat and rice, its design principles can be adapted to monitor a wide variety of crops, making SMICGS a valuable tool for global food security.
###
References
DOI
10.1016/j.plaphe.2025.100056
Original URL
https://doi.org/10.1016/j.plaphe.2025.100056
Funding information
This work was supported by the National Key Research and Development Program of China (2021YFD2000101) and Jiangsu Province Science Foundation for Youths (BK20241544).
About Plant Phenomics
Plant Phenomics is dedicated to publishing novel research that will advance all aspects of plant phenotyping from the cell to the plant population levels using innovative combinations of sensor systems and data analytics. Plant Phenomics aims also to connect phenomics to other science domains, such as genomics, genetics, physiology, molecular biology, bioinformatics, statistics, mathematics, and computer sciences. Plant Phenomics should thus contribute to advance plant sciences and agriculture/forestry/horticulture by addressing key scientific challenges in the area of plant phenomics.