Estimating leaf chlorophyll content by laser-induced fluorescence technology at different viewing zenith angles

Area of Science:

• Plant Physiology and Remote Sensing
• Spectroscopy and Analytical Chemistry

Background:

• Leaf chlorophyll content (LCC) is crucial for assessing plant health and growth.
• Non-destructive, remote sensing methods are vital for efficient plant monitoring.
• Laser-induced fluorescence (LIF) offers a promising technique for LCC estimation.

Purpose of the Study:

• To compare the predictive accuracy of various single and combined fluorescence characteristics for LCC estimation using LIF.
• To evaluate the influence of different viewing zenith angles (VZAs) on LCC prediction accuracy.
• To identify optimal multivariate analysis algorithms and input variables for robust LCC monitoring.

Main Methods:

• Developed support vector machine (SVM) models for LCC estimation.
• Utilized fluorescence characteristics (fluorescence peak, fluorescence ratio, PCA, first-derivative) and their combinations as input variables.
• Investigated various viewing zenith angles (0°, 15°, 30°, 45°, 60°) for spectral data acquisition.

Main Results:

• The fluorescence ratio (FR) demonstrated superior predictive performance among single fluorescence characteristics.
• SVM models utilizing combined fluorescence characteristics (FP+FR+FD) showed enhanced LCC estimation accuracy.
• Optimal LCC monitoring was achieved at a 0° viewing zenith angle, with 0° and 30° yielding superior results compared to other angles.

Conclusions:

• The combination of LIF technology and multivariate analysis, particularly SVM with optimized fluorescence characteristics and VZAs, provides an effective method for LCC monitoring.
• Viewing zenith angle significantly impacts LCC estimation accuracy, highlighting the need for careful selection.
• This approach holds significant potential for accurate plant growth diagnosis and assessment.