Leaf diseases pose a significant threat to agricultural productivity and ecosystem stability worldwide. This paper employs an advanced CNN model—Efficient Net, to accurately perceive and classify the fungal leaf contagions in tomato and grape plants, such as late blight and early blight in tomatoes and black rot, esca, grape blight in grapevines. The model provides a precise calculation of disease spread percentage on each leaf, offering valuable discernments. This quantification allows for a more accurate assessment of infection severity, enabling targeted disease management strategies. The results offer a detailed quantitative analysis of leaf infirmities, enhancing the understanding of disease progression in grape and tomato plants, with the model achieving 89% accuracy. Thus, the study clearly depicts the plant contagions in Grape and Tomato leaves with a comprehensive quantitative analysis of leaf diseases that promotes health and ecosystem resilience in flora life.