Non-Destructive Estimation of Total Chlorophyll Content of Apple Fruit Based on Color Feature, Spectral Data and the...

Area of Science:

• Agricultural Science
• Food Science
• Spectroscopy

Background:

• Non-destructive assessment of fruit quality is crucial for various stages from growth to market.
• Physicochemical properties, like chlorophyll content, indicate internal fruit quality.

Purpose of the Study:

• To estimate total chlorophyll content in Fuji apples non-destructively.
• To utilize color and spectral data for accurate chlorophyll prediction.
• To develop and validate hybrid artificial intelligence algorithms for this purpose.

Main Methods:

• Extraction of spectral and color data from Fuji apple samples.
• Identification of optimal wavelengths (660-720 nm) using Artificial Neural Network-Particle Swarm Optimization (ANN-PSO).
• Prediction of total chlorophyll content using Artificial Neural Network-Imperialist Competitive Algorithm (ANN-ICA) with spectral and color data.

Main Results:

• High accuracy in predicting total chlorophyll content was achieved.
• The best prediction models, using spectral data and optimal wavelengths, showed R² values up to 0.997 and RMSE as low as 0.0006.
• The hybrid ANN-ICA demonstrated reliability through 1000 iterations.

Conclusions:

• Non-destructive estimation of total chlorophyll content in apples is feasible using spectral and color data.
• Hybrid AI algorithms, particularly ANN-ICA, are effective tools for predicting fruit quality parameters.
• This method offers a valuable approach for real-time fruit quality monitoring.