Web-Based Application for Plant Leaf Disease Detection

Abstract

Several diseases that affect plant leaves pose a severe hazard to agriculture. Our approach identifies both the disease that harmed the leaf and the area of harm. Crop diseases, especially those that predominantly affect the leaves, have an impact on both the quantity and quality of agricultural output. The objective of the article is to raise consciousness among farmers about the latest innovations that can prevent disease of plant leaves. The techniques of data mining and image processing with an accurate algorithm have been identified to detect leaf illnesses in the potato plant as potatoes are only an easily accessible vegetable. In this study, we present a web-based automated approach for identifying and classifying plant leaf diseases. The recommended approach involves receiving an input image, delivering it to the model using a Postman API, analyzing the image using a CNN model kept inside a Docker container, and producing a result to determine if the image is classified as healthy or unhealthy. The Plant Village dataset for plants like tomatoes and potatoes is used to validate this study. The accuracy of the proposed model is 98.44% on potato leaf samples.