The health of plants is crucial for agricultural productivity and food security. Early detection of plant diseases can significantly mitigate the impact of diseases on crop yields. This paper presents a novel plant disease detection system that leverages machine learning and image processing techniques to identify and diagnose plant diseases. The system utilizes a convolutional neural network (CNN) to analyse images of plant leaves, distinguishing between healthy and diseased specimens. By employing a comprehensive dataset of plant images, the model is trained to recognize various diseases and provide accurate diagnostics. The system offers real-time detection capabilities, which can be integrated into mobile applications for field use. This approach aims to enhance disease management practices, reduce reliance on manual inspections, and ultimately support sustainable agricultural practices.