Skip to content

UAVVaste: COCO-like dataset and effective waste detection in aerial images

License

Notifications You must be signed in to change notification settings

PUTvision/UAVVaste

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

UAVVaste dataset

The UAVVaste dataset consists to date of 772 images and 3718 annotations. The main motivation for the creation of the dataset was the lack of domain-specific data. Therefore, this image set is recommended for object detection evaluation benchmarking but also for developing solutions related to UAVs, remote sensing, or even environmental cleaning. The dataset is made publicly available and will be expanded.

Date Images count Annotations count
14.11.2020 772 3718

The dataset is a part of the research paper: ,,Autonomous, Onboard Vision-Based Trash and Litter Detection in Low Altitude Aerial Images Collected by an Unmanned Aerial Vehicle'' Marek Kraft, Mateusz Piechocki, Bartosz Ptak and Krzysztof Walas - MDPI Remote Sensing.

Usage

Requirements

pip3 install -r requirements.txt

Download

UAVVaste dataset is now available also in the Zenodo repository: https://zenodo.org/record/8214061.

python3 main.py

Contribute

If you want to contribute to the UAVVaste dataset, please contact us via email [email protected]. We will be grateful for all images taken from the UAV with litter annotations in the COCO format (bounding boxes and segmentation masks).

Citation

@Article{rs13050965,
AUTHOR = {Kraft, Marek and Piechocki, Mateusz and Ptak, Bartosz and Walas, Krzysztof},
TITLE = {Autonomous, Onboard Vision-Based Trash and Litter Detection in Low Altitude Aerial Images Collected by an Unmanned Aerial Vehicle},
JOURNAL = {Remote Sensing},
VOLUME = {13},
YEAR = {2021},
NUMBER = {5},
ARTICLE-NUMBER = {965},
URL = {https://www.mdpi.com/2072-4292/13/5/965},
ISSN = {2072-4292},
DOI = {10.3390/rs13050965}
}