PLANT LEAF DISEASE DETECTION SYSTEM USING CNN (KEY IJP************632)

• Anjali Panchal,Swamini Zinje,Shrawani Wadhwane,Prof. Sayali Karmode

Abstract

Abstract-Plant diseases pose a continuous danger to smallholder farmerslivelihoods and food security. Images I agriculture may now be classified using computer vision models thanks to the recent revolution in smartphone penetration. In terms of image recognition, convolutional neural networks (CNNs) are thought to be state-of-the-art because they can quickly and definitively diagnose a condition. This paper examines the efficacy of a pre-trained ResNet34 model in identifying agricultural diseases. The created model can identify seven plant illnesses from healthy leaf tissue and is available as a web application. A dataset including photographs of leaves, taken under controlled conditions, is created for the purpose of training and verifying the model. The suggested approach can attain an accuracy of 97.2% and an F1 score of more than 96.5%, according to validation results. This shows that CNNs can classify plant illnesses technically and offers a way forward for small-holder farmers to use AI solutions.