This record provides the Oil Tank Detection Dataset presented in: 1) Rizk, M., & Chehade, A., “Efficient Oil Tank Detection Using Deep Learning: A Novel Dataset and Deployment on Edge Devices”, IEEE Access, 2024. DOI: 10.1109/ACCESS.2024.3495523.2) Chehade, A., & Rizk, M., “Unveiling a Cutting-Edge Dataset for Oil Tank Detection: YOLO Models Put to the Test”, 2024 International Conference on Smart Systems and Power Management (IC2SPM), 2024. DOI: 10.1109/IC2SPM62723.2024.10841353. The dataset contains 12,948 satellite images and 171,809 bounding-box annotations for oil storage tank detection. It combines (i) 2,090 high-resolution Google Earth images (1105×720) manually annotated using CVAT, and (ii) three publicly available datasets that were converted to YOLO format, refined, and expanded with improved annotations. Annotations follow the YOLO bounding-box format and provide a single target category (oil storage tank). The dataset is intended for training and evaluating object detection methods in remote sensing and infrastructure monitoring. Code and repository:https://github.com/adelchehade99/Oil-Tank-Detection