DeepSpaceYoloDataset: an annotated set of smart telescopes images

During the MILAN research project (MachIne Learning for AstroNomy), we have compiled a large collection of deep sky images during Electronically Assisted Astronomy sessions in Luxembourg, France, Belgium.

We have used two instruments for several months (from March 2022 to September 2023): a Stellina smart telescope (https://vaonis.com/stellina) and a Vespera smart telescope (https://vaonis.com/vespera).
We have captured data for a representative set of deep sky objects from the Messier / NGC / IC / Sharpless2 / Barnard catalogues.
Different types of celestial objects were considered: emission/reflection/dark/planetary nebula, galaxies, globular/open clusters.
Images were obtained after the capture and the stacking of sub-frames of 10 seconds exposure time.
Training images were splitted into 608x608 patches.
Based on the YOLOv7 format, the dataset is a ZIP file containing 4696 RGB images, and the corresponding 4696 labels text files with the positions of deep sky objets in the images. 

In 2026, we expanded the dataset with a new set of 335 high-resolution images obtained by stacking multiple nights of observations conducted between March 2022 and April 2026. 
These higher-quality images were manually annotated and then added to the test2026 directory, enabling a more in-depth evaluation of  deep sky objects detection models.

This research was initially funded by the Luxembourg National Research Fund (FNR), grant reference 15872557.

More information about the MILAN project: https://www.fnr.lu/results-2021-1-bridges-call/.
More information about the MILAN2 project: https://researchportal.list.lu/projects/detail/milan2.

More information about VAONIS instruments: https://vaonis.com

More information about Luxembourg of Science and Technology (LIST): https://www.list.lu

Data license for files: Attribution-NonCommercial-NoDerivatives 4.0 International