Banana leaf spot diseases such as Cordana, Pestalotiopsis, and Sigatoka pose significant challenges to the banana sector worldwide, limiting the supply of essential nutrients as well as the stability of the global economy. In this study, we utilize deep learning (DL) models to ensure the precise detection of these diseases in a proper way with better accuracy. Currently, detection methods for these diseases are often labor-intensive, time-consuming, and tending to inaccuracies. A significant gap exists in the development of efficient and precise detection techniques that can be readily deployed in the field. Our research emphasizes the importance of effective management techniques to ensure food security and long-term livelihood. With a subset of the Banana Leaf Spot Disease (BananaLSD) dataset, the experimental findings with YOLOv8 showed promising metrics: a precision rate of 88.9%, and a recall rate of 92%, for the YOLOv8s model. These results have the potential to significantly increase crop yields and allow for more accurate leaf spot disease detection in fields.