Volume 1 - Issue 4, November - December 2025

As a staple crop, potatoes are essential to the world's food security. Thus, this effort aims to explore CNNs' potential for potato leaf disease detection. To efficiently classify different potato leaf diseases, we use a Convolutional Neural Network (CNN) technique. The CNN model uses several convolutional, pooling, and fully connected layers to extract spatial characteristics from the input images. Many academics have sought to improve the early detection of potato blight using various machine and deep learning techniques as neural networks have been included into agriculture. By learning hierarchical cues including color variations, texture, and form patterns, the network effectively differentiates between healthy and sick leaves. Additionally, the CNN model records 96.52% accuracy, 96.67% precision, 96.52% recall, and an F1 score of 96.52% for severity categorization, demonstrating its capacity to handle intricate visual patterns in agricultural disease detection.