This study addresses the significant impact of leaf diseases, such as rust and blast, on the agricultural productivity of pearl millet, a crop known for its high nutritional value and micronutrient content. In tackling these challenges, the re-search introduces a deep learning-based approach for the accurate and minimally supervised identification and diagnosis of these diseases. The methodology employs advanced deep learning algorithms, including VGG-16, Mobile Net V1, and Mobile Net V2, renowned for their pattern recognition capabilities and adapt-ability in applying knowledge from previous tasks to new scenarios. These models are specifically utilized to detect rust and blast diseases in pearl millet. Performance metrics such as accuracy, precision, recall, and F1 score are used to evaluate the effectiveness of these models. Results indicate a high level of accuracy, with the VGG-16 model achieving 99.45%, and both Mobile Net V1 and Mobile Net V2 models showing an accuracy of 99.32% in detecting diseased leaves. This research not only contributes to advancements in agricultural technology but also provides valuable tools for farmers and the agricultural industry to manage crop diseases more efficiently.