PotatoGuard: A Robust System for Potato Disease Automated Detection and Classification

Grown as a major food crop throughout the world, potatoes, ranks fourth in importance among all crops that humans consume after wheat, rice, and maize. Carbohydrates, vitamins, and minerals are abundant in potatoes and this makes them an essential food source for a large number of people. Potatoes are not just nutrient-dense but also highly versatile. It can be utilized in a variety of recipes, from eating raw food to processed foods like fries, chips, and carbs. Nonetheless, diseases like early and late-blight are frequently observed in its leaves. Food security could be impacted by this, and there might be significant output losses. Consequently, management of this is critical to maintaining potato productivity and ensuring sustainable agricultural techniques. This paper's primary goal is to develop a system for classifying potato diseases using machine learning techniques like CNN (Convolutional Neural Network). This study proposes a thorough method for categorizing potato illnesses through machine learning approaches. This study makes use of a large data set that includes pictures of potato leaves with various illnesses, such as "early blight" and "late blight." Following training and validation on multiple preprocessing layers in the dataset, the model demonstrated a high degree of accuracy in differentiating between healthy and unhealthy potato leaves. Additionally, the model was enhanced by rotating and altering the contrast of the photographs using data augmentation techniques. The results provide precise and efficient methods for detecting potato diseases, which is helpful for early disease diagnosis and mitigation strategies.