An interpretable crop leaf disease and pest identification model based on prototypical part network and contrastive learning

Nature — Machine LearningTuesday, November 4, 2025 at 12:00:00 AM
A new model for identifying crop leaf diseases and pests has been developed using prototypical part networks and contrastive learning. This innovative approach not only enhances the accuracy of disease detection but also makes the process interpretable for farmers and agricultural experts. By improving the ability to diagnose issues in crops, this model can lead to better crop management and increased yields, which is crucial for food security and sustainable agriculture.
— via World Pulse Now AI Editorial System

Was this article worth reading? Share it

Recommended apps based on your readingExplore all apps
Continue Readings
Antimicrobial peptide prediction based on contrastive learning and gated convolutional neural network
NeutralArtificial Intelligence
A recent study published in Nature — Machine Learning introduces a novel approach for predicting antimicrobial peptides using contrastive learning and gated convolutional neural networks. This method aims to enhance the accuracy and efficiency of peptide prediction, which is crucial for developing new antimicrobial agents.