Smart Multi Fruit Disease Detection and Pesticide Recommendation System

Fruit diseases are major challenge to the agricultural sector and lead to low yields or poor-quality crops, which again affects their market value as well as the income that farmers may get from those crops. Effective control requires quick identification of the causative organism so that it can be targeted with narrow-spectrum pesticides rather than broadspectrum ones and hence helps in both economic and environmental sustainability. This paper is reporting a web- based system that classifies multiple fruit diseases and provides the best pesticide treatments by using machine learning. Its key component has been a deep learning model, which has been trained using an exhaustive dataset of fruit photos; the pictures contained several states of disease in important fruits like guava, orange, papaya, and mango. The system dynamically generates customized pesticide recommendations in line with best practices for safe and effective use, and it is very accurate in disease detection.