Agriculture is a major part of many nations’ economies, and in Telangana, it plays a crucial role. Rice, also referred to as paddy is cultivated as a staple food cash crop, as rice is consumed three times a day by the people of Telangana. However, various factors such as diseases affecting paddy leaves, temperature fluctuations, and pest attacks can significantly impact paddy production, potentially reducing yields by 40–50%. The prevalence of paddy diseases varies by region and season, with the most common ones in Telangana being bacterial leaf blight, brown spot, and rice blast. Early detection of these diseases is essential to prevent widespread damage to farmland. We address this through our proposed system that uses preprocessing and data augmentation to expand the dataset. It employs a pretrained Inception_v3 algorithm and a deep learning-based CNN algorithm, which efficiently classifies paddy diseases. This approach not only enhances the accuracy of disease prediction but also reduces the time required for classification, helping farmers take timely action to protect their crops.