Deployable lightweight deep learning models for multiclass classification of cocoa and coffee seeds

Key Points

• The aim is to develop and evaluate deployable deep learning models for classifying cocoa and coffee seeds into multiple categories.
• Developed lightweight deep learning models for classification
• Applied models to both cocoa and coffee seed datasets
• Conducted evaluations of model performance in multiclass scenarios
• Achieved high accuracy in classifying both cocoa and coffee seeds
• Models demonstrate efficient processing with minimal computational resources
• Highlighted improvements in classification speed compared to traditional methods