Images of Recyclable Waste Dataset
This open dataset contains multiple images of recyclable waste on a conveyor belt and it is produced as part of the EU funded project RECLAIM: GA-101070524.
The exploitation plan of RECLAIM is centered on the open-source distribution of project results to accelerate developments on smart, automated material recovery. In this framework, the dataset developed as part of the RECLAIM project is available as open source, for re-use and potential advancement by researchers and developers.
This dataset consists of several folders, one for each recyclable material type (PET, HDPE, aluminum, PP, PS, TetraPak, and a mixed-type) which are subdivided in two other folders for RGB and HSI images respectively. The dataset of RGB waste images was collected and curated by FORTH (Institute of Computer Science), while the dataset of hyperspectral waste images was collected and curated by IRIS Technology Solutions.
RECLAIM's main objective for material recognition is the combination of imaging in the visual and infrared domain to identify, localize and categorize recyclables.