[Hyperspectrum based prediction model for nitrogen content of apple flowers]

Area of Science:

• Agricultural Science
• Remote Sensing
• Plant Nutrition

Background:

• Accurate nitrogen content assessment is crucial for apple cultivation management.
• Traditional methods for nitrogen analysis are often time-consuming and labor-intensive.

Purpose of the Study:

• To quantitatively retrieve nitrogen content in apple flowers using hyperspectral imaging.
• To establish a basis for informatized apple management and nutrition diagnosis.

Main Methods:

• Collected hyperspectral reflectance data from 120 apple flower samples.
• Analyzed correlations between spectral characteristics (original and first derivative) and nitrogen content.
• Developed and optimized prediction models using sensitive spectral bands.

Main Results:

• Identified significant positive and negative correlations between spectral reflectance and nitrogen content across various wavelength ranges.
• Selected original spectral reflectance at 640 nm and 676 nm for optimal prediction models.
• Achieved high prediction accuracy, with R² values of 0.8258 and 0.8936, and average accuracies of 92.9% and 94.0%.

Conclusions:

• Hyperspectral modeling provides an effective and rapid method for predicting apple flower nitrogen content.
• The developed models offer theoretical basis and technical support for precision agriculture in apple orchards.
• This approach facilitates timely nutrient diagnosis and management decisions.