We're excited to share the latest developments on Maize Watch, our corn disease detection platform! Our team has successfully implemented an enhanced ResNeXt model that integrates dual attention mechanisms - Efficient Channel Attention (ECA) and Spatial Attention - significantly improving detection accuracy across various corn diseases including Common Rust, Gray Leaf Spot, and Northern Leaf Blight. A major breakthrough this week was our implementation of Grad-CAM visualization tools that make our AI's decision-making process transparent. Farmers can now see exactly which parts of a leaf image triggered a disease diagnosis, building trust in the technology. The updated model achieves over 95% accuracy on our test dataset while maintaining fast inference times suitable for mobile devices. Look for these improvements in our next app store release coming later this month!
Log in or sign up for Devpost to join the conversation.