Agricultural products are essential for every country. When plants become diseased, they negatively impact agricultural production and the country’s economy. This paper aims to propose a technique for identifying and detecting plant leaf diseases using deep learning techniques. The images used were sourced from the Plant Village dataset. This study focuses on categorizing plants, specifically tomatoes, peppers, and potatoes, among the many plant types globally. The dataset includes 20,638 images of plants and their diseases. The proposed system uses a Convolutional Neural Network (CNN) to classify plant leaf diseases into 15 categories: 12 classes for various diseases, such as bacterial and fungal infections, and 3 classes for healthy leaves. By addressing this issue, this study strives to achieve high accuracy in disease classification and provides a valuable tool for the early detection and management of agricultural diseases. As a result, the system achieved high accuracy, with 95.62% for training and 96.14% for testing across all datasets used.