[Study of building quantitative analysis model for chlorophyll in winter wheat with reflective spectrum using MSC-ANN...

Area of Science:

• Agricultural Science
• Spectroscopy
• Data Analysis

Background:

• Environmental factors introduce noise into spectral data, complicating accurate analysis.
• Predicting chlorophyll content in winter wheat is crucial for agricultural management.

Purpose of the Study:

• To develop and validate an accurate and efficient method for predicting chlorophyll content in winter wheat using near-infrared spectroscopy.
• To effectively mitigate spectral noise caused by environmental factors.

Main Methods:

• Multiplicative Scatter Correction (MSC) was applied to preprocess spectral data.
• Principal Component Analysis (PCA) using Nonlinear Iterative Partial Least Squares (NIPALS) identified key spectral components.
• Back Propagation Artificial Neural Networks (BP-ANN) were trained to model the relationship between spectral data and chlorophyll content.

Main Results:

• The developed MSC-ANN model demonstrated high accuracy, with correlation coefficients (r) of 0.9604 for the calibration set and 0.9600 for the prediction set.
• Low standard deviation (SD) and relative standard deviation (RSD) values indicate the model's precision in predicting chlorophyll content.
• The method effectively removed spectral noise, enabling precise chlorophyll content prediction in living leaves.

Conclusions:

• The MSC-ANN algorithm provides an effective and accurate method for predicting chlorophyll content in winter wheat.
• This approach offers a reliable alternative to classical methods, meeting the demand for rapid agricultural product analysis.
• The study highlights the potential of spectral analysis combined with advanced algorithms for precision agriculture.