Development of a Target-to-Sensor Mode Multispectral Imaging Device for High-Throughput and High-Precision...

Area of Science:

• Agricultural Science
• Plant Biology
• Spectroscopy

Background:

• Image-based spectroscopy phenotyping analyzes genotype-environment-management interactions using light wavelengths.
• Remote sensing faces noise from lighting and plant status, while proximal methods are labor-intensive.
• Current proximal leaf-scale imaging requires manual sample handling, increasing time and cost.

Purpose of the Study:

• To develop a proximal multispectral imaging device for high-precision, high-throughput leaf-scale phenotyping.
• To overcome limitations of existing remote and proximal sensing techniques in plant phenotyping.
• To introduce a novel target-to-sensor approach using active airflow for leaf repositioning and stabilization.

Main Methods:

• Developed a proximal multispectral imaging device employing active airflow to attract and flatten leaves (target-to-sensor mode).
• Utilized artificial lighting to ensure stable and consistent imaging conditions, minimizing ambient light interference.
• Compared the device's performance against traditional remote sensing systems in field tests.

Main Results:

• Achieved a five-fold increase in imaging speed compared to conventional proximal devices.
• Successfully identified nitrogen deficiency at an earlier stage than remote sensing systems.
• Demonstrated significantly lower p-values (0.008) for data collected by the new device versus remote sensing (0.239).

Conclusions:

• The novel target-to-sensor proximal imaging device significantly enhances throughput and accuracy in leaf-scale plant phenotyping.
• Active airflow mechanism effectively repositions and flattens leaves, optimizing image quality and reducing operator labor.
• The device offers superior sensitivity for early stress detection, outperforming traditional remote sensing methods.