Integrated hyperspectral–RGB modeling for the estimation of wheat protein and dough rheological properties using machine and deep transfer learning

Key Points

• Wheat protein and dough rheological properties are effectively estimated using integrated hyperspectral and RGB modeling.
• The study demonstrates a high predictive accuracy of 92% for protein content and 88% for dough properties based on machine learning algorithms.
• Analysis employed deep transfer learning methods to enhance model performance and reliability across diverse datasets.
• This innovative approach highlights the potential for advanced agricultural monitoring and precision farming strategies.