Research on Tobacco Leaf Color Classification Based on Machine Learning

Objective: This study explores the application of machine learning algorithms in tobacco leaf color classification to improve classification accuracy.Methods: Images of tobacco leaves of different grades were collected, and RGB color data was extracted. Six machine learning algorithms were employed for classification prediction, and the best algorithm and features were optimized.Results: After denoising, algorithm optimization, and feature introduction, the classification accuracy of the random forest algorithm improved from 75% to 90.77%.Conclusion: Machine learning effectively achieves automatic classification of tobacco leaf colors, with data quality, algorithm selection, and feature engineering significantly impacting model performance. This research provides technical support for intelligent grading of tobacco leaves, though attention must be given to the model's generalization ability and processing speed.