Detection and Classification of Saffron Plant Diseases using Machine Learning: A Research

Abstract

Abstract: Saffron, known as "red gold," is a valuable spice derived from the flower of Crocus sativus. However, saffron cultivation faces challenges due to diseases that can harm crop yield and quality. This thesis proposes a deep learning-based approach using the VGG 16 architecture to detect and classify saffron diseases. The study collects a comprehensive saffron disease dataset, organizes it by disease type, and enhances its quality through analysis and augmentation. The VGG 16 architecture, known for image classification, is adapted for saffron disease detection, utilizing convolutional and fully connected layers for feature extraction and classification. The model is trained using multiple epochs, achieving an impressive 87% accuracy. Comparison with other methods demonstrates the superiority of the proposed approach. The study utilizes highperformance computing systems for efficient evaluation. Overall, this research demonstrates the potential of deep learning in saffron disease management, aiding farmers in effective decision-making for disease control measures.

Code and data availability

The paper describes a VGG-16 saffron disease classifier trained on images collected from a saffron research center, but no public dataset, image repository, code, or trained model is deposited or linked anywhere in the supplied blocks. No availability statement or authors' public URL exists.

No evidence-backed public reproduction asset is currently recorded.