Potato cultivation has been prevalent in India for decades. However, due to the same potato plant diseases, potato cultivation needs to be improved. Based on the leaf condition, the potato plant disease can be categorized into early blight, late blight, etc. This research aims to develop an accurate and efficient system for identifying and classifying diseases by analyzing digital images of the leaves. A dataset of high-resolution images of healthy and diseased potato leaves is used as the foundation for training and evaluating the machine-learning models. The methodology involves implementing state-of-the-art deep learning architectures, including Convolutional Neural Networks (CNNs) and transfer learning techniques. Pre-trained models are leveraged to enhance the efficiency of the disease detection system. Results of the performance metrics indicate the potential for practical implementation in agricultural settings.