Hyperspectral imaging for estimating leaf, flower, and fruit macronutrient concentrations and predicting strawberry...

Area of Science:

• Agricultural Science
• Remote Sensing
• Plant Nutrition

Background:

• Optimizing strawberry yield and quality relies on effective management of plant nutritional status.
• Accurate assessment of nutrient concentrations (nitrogen, phosphorus, potassium, calcium) is crucial for informed agricultural practices.

Purpose of the Study:

• To evaluate hyperspectral imaging for estimating nutrient concentrations in various strawberry plant parts (leaves, flowers, unripe and ripe fruit).
• To assess the potential of hyperspectral imaging in predicting strawberry plant yield.
• To determine the accuracy of partial least squares regression (PLSR) models for nutrient estimation.

Main Methods:

• Hyperspectral imaging was employed across the 400-1,000 nm spectrum.
• Partial least squares regression (PLSR) models were developed to correlate spectral data with nutrient concentrations.
• Prediction accuracy was evaluated using the coefficient of determination (R²ₚ) and ratio of performance to deviation (RPD).

Main Results:

• Hyperspectral imaging demonstrated good prediction accuracy for nitrogen and calcium concentrations, particularly in leaves and flowers.
• Prediction accuracy was generally higher for leaves, flowers, and unripe fruit compared to ripe fruit.
• Yield and fruit mass showed limited linear relationships with tested vegetation indices, with the Difference Vegetation Index being significant.

Conclusions:

• Hyperspectral imaging is a promising technology for non-destructively assessing nutrient status in strawberry crops.
• This approach can support growers in making rapid, data-driven nutrient management decisions.
• Implementing hyperspectral imaging can lead to optimized strawberry yield and improved fruit quality.