Underwater Drowning Detection DatasetThis dataset contains 5,613 manually annotated underwater images for drowning detection research, captured in controlled swimming pool environments. It provides a balanced distribution of three behavioral states:Swimming (1,871 images)Struggling (1,871 images)Drowning (1,871 images)All images were collected under real underwater conditions and annotated for object detection tasks using the YOLO format.Key FeaturesHigh-resolution underwater images (640×640 pixels, RGB)YOLO .txt annotations with bounding boxes for three behavior classesBalanced class distribution to minimize model biasData collected ethically with lifeguard supervision and participant consentIncludes realistic challenges such as water distortion and lighting variabilityTechnical DetailsTotal Images: 5,613Training/Validation Split: 4,488 / 1,125Classes: Swimming, Struggling, DrowningFormat: JPEG + YOLO annotation filesResolution: 640×640 pixelsBaseline Performance: YOLOv8n achieved 97.5% mAP@50 on this datasetAnnotation FormatEach image has a corresponding .txt file with annotations in YOLO format, where each line follows this structure: Field Descriptions:class_id: Integer label for the class0 = Swimming1 = Struggling2 = Drowningx_center, y_center: Normalized center coordinates of the bounding box (values between 0.0 and 1.0)width, height: Normalized width and height of the bounding box (values between 0.0 and 1.0)Example Annotation:0 0.509896 0.568519 0.453125 0.581481This line indicates a “Swimming” detection (class_id = 0) with a bounding box centered at 50.99% (horizontal) and 56.85% (vertical) of the image dimensions, covering 45.31% of the width and 58.15% of the height.Dataset Folder Structuredatasets/├── images/│ ├── train/│ │ ├── frame_00001.jpg│ │ └── ...│ └── val/│ ├── frame_04489.jpg│ └── ...│├── labels/│ ├── train/│ │ ├── frame_00001.txt│ │ └── ...│ └── val/│ ├── frame_04489.txt│ └── ...│├── classes.txt├── README.mdUse and ApplicationsThis dataset is designed to support the development and evaluation of real-time AI systems for aquatic safety, including:Drowning detection modelsMulti-class object detection in underwater environmentsResearch in underwater computer vision and human activity recognitionCitationIf you use this dataset, please cite:graphqlCopyEdit@dataset{underwater_drowning_detection_2025, title = {Underwater Drowning Detection Dataset}, author = {Hamad Alzaabi and Saif Alzaabi and Sarah Kohail}, year = {2025}, publisher = {Figshare}, note = {Manually annotated underwater images for drowning detection research}}Please also cite the related publication:mathematicaCopyEdit@inproceedings{Alzaabi2025, author = {Hamad Alzaabi and Saif Alzaabi and Sarah Kohail}, title = {Multi‑Swimmer Drowning Detection Using a Custom Annotated Underwater Dataset and Real‑Time AI}, booktitle = {Proceedings of the International Conference on Image Analysis and Processing (ICIAP)}, year = {2025}}

Dataset

This dataset was created by Alanoud Awaji

Contents

An issue of growing importance within the field of drowning prevention is the undertaking of aquatic rescues by bystanders, who sometimes drown in the process. The main objectives of this study were to describe characteristics of bystanders making rescues in different Australian aquatic environments, identify the role of prior water safety training in conducting bystander rescues and provide insights into future public education strategies relating to bystander rescue scenarios. An online survey was disseminated via various social media platforms in 2017 and gathered a total of 243 complete responses. The majority of bystander rescues described took place in coastal waterways (76.5%; n = 186), particularly beaches (n = 67), followed by pools (17.3%; n = 42) and inland waterways (6.2%; n = 15). The majority of respondents were males (64.2%; n = 156) who rescued on average approximately twice as many people in their lifetime (6.5) than female respondents (3.6). Most rescues occurred more than 1 km from lifeguard/lifesaver services (67%; n = 163), but in the presence of others (94.2%; n = 229). The majority of bystander rescuers had water safety training (65.8%; n = 160), self-rated as strong swimmers (68.3%; n = 166), conducted the rescue without help from others (60%; n = 146), did not use a flotation device to assist (63%; n = 153), but were confident in their ability to make the rescue (76.5%; n = 186). However, most considered the situation to be very serious (58%; n = 141) and felt they had saved a life (70.1%; n = 172). With the exception of pools, most bystanders rescued strangers (76.1%; n = 185).While Australia clearly benefits from having a strong water safety culture, there is no clear consensus on the most appropriate actions bystanders should take when confronted with a potential aquatic rescue scenario. In particular, more research is needed to gather information regarding bystander rescues undertaken by those without prior water safety training.