Detection and Classification of Plant Disease Using Artificial Intelligence

Grapevine, a vital crop in India, faces substantial threats from diseases influenced by both poor soil quality and unfavorable weather conditions. The economic impact on farmers due to crop damage is significant. To mitigate these losses, an early detection system is crucial. This paper proposes a computer vision approach, leveraging computer vision and convolutional neural networks (CNNs), specifically designed for grapevine health assessment. The study employs four pre-trained CNN architecture-AlexNet, GoogleNet, ResNet and SqueezeNetusing transfer learning on dataset comprising images of three diseased grapevine leaves. The classification is achieved through a fully connected layer, and the results indicate that ResNet-18 attains the highest accuracy, demonstrating its effectiveness in identifying diseases in grapevine crops. This approach holds promise for enhancing grapevine crop management and reducing the economic burden on farmers.