Crop production worldwide is adversely affected by various plant diseases, highlighting the importance of disease identification and classification. Traditionally, this process relied on specialists and experts, which proved to be time-consuming and cumbersome. However, advancements in machine learning and deep learning have revolutionized disease classification. Convolutional Neural Networks (CNNs), particularly in the domain of plant leaf diseases, have demonstrated remarkable success. This paper focuses on training pre-trained models using three distinct plant leaf disease datasets—Mango, Guava, Black gram, and Sugarcane with varying parameters to identify the most effective models achieving high accuracy. Additionally, the study explores the impact of data augmentation, fine-tuning, and dropout techniques on model performance. The findings demonstrate the effectiveness of transfer learning in plant disease identification, with fine-tuned CNN models achieving remarkable accuracy.