Optimized deep learning with Grad-CAM for automated cardamom classification: A multispectral imaging approach for real-time mobile deployment | Synapse
March 3, 2026
Optimized deep learning with Grad-CAM for automated cardamom classification: A multispectral imaging approach for real-time mobile deployment
Key Points
Automated classification accuracy is significantly improved through optimized deep learning techniques—demonstrating a robust approach for cardamom.
Key evidence shows that the implementation of Grad-CAM resulted in a classification accuracy of over 95% across various samples.
Observational analysis utilizing multispectral imaging establishes a fast and efficient method for real-time classification in mobile settings.
This work highlights potential advancements in agricultural technology, especially for field-level applications.