Management of crop health and guarantes of food security depend on accurate detection of maize leaf diseases. highly laborious and prone to errors are conventional manual inspection techniques. This work uses a convolutional neural network can automatically increase disease detection precision for maize leaves. With images scaled to $256 \times 256$ pixels and enhanced through rotation and scaling to boost generalization, the CNN model was Instruction utilising a photo dataset. classified into four categories: Common rust; blight; grey leaf spot; healthy leaves. Built with convolutional, dropout, pooling, fully connected layers, trained with a batch size of 32 using categorical cross-entropy loss and Adam optimiser spanning 15 epochs. With an overall accuracy of 89%, evaluation guidelines comprising recall, precision and F1-score point to excellent performance. While Blight (precision: 0.84, recall: 0.87) and Grey Leaf Spot (precision: 0.73, recall: 0.71) exhibit somewhat less performance, the model shines in spotting Healthy leaves (precision: 1.00, recall: 0.99) and Common Rust (precision: 1.00, recall: 0.99). With reliable disease diagnosis to promote crop management and enhance food security, the CNN model has great practical relevance in agriculture.
PlantPulse: Intelligent CNN-based Analysis for Effective Maize Leaf Disease Detection
2024-11-06
445700 byte
Conference paper
Electronic Resource
English
Leaf Disease Detection Using Deep Learning
Springer Verlag | 2024
|A Weakly Supervised Approach for Disease Segmentation of Maize Northern Leaf Blight from UAV Images
DOAJ | 2023
|