GlobalWasteData: A Large-Scale, Integrated Dataset for Robust Waste Classification and Environmental Monitoring

TL;DR

GlobalWasteData addresses the need for a robust, generalizable waste classification benchmark by integrating 19 public datasets into a single, large-scale archive with 89,807 images, 14 main categories and 68 subclasses. The authors establish a unified preprocessing pipeline, consistent directory-based annotations, and a stratified train/validation/test split to reduce label drift and domain bias. Key contributions include data harmonization, standardized labeling, licensing transparency under CC BY 4.0, and public availability to foster reproducibility and cross-domain research. The dataset aims to improve model generalization across diverse real-world waste environments and facilitate environmental monitoring and recycling automation.

Abstract

The growing amount of waste is a problem for the environment that requires efficient sorting techniques for various kinds of waste. An automated waste classification system is used for this purpose. The effectiveness of these Artificial Intelligence (AI) models depends on the quality and accessibility of publicly available datasets, which provide the basis for training and analyzing classification algorithms. Although several public waste classification datasets exist, they remain fragmented, inconsistent, and biased toward specific environments. Differences in class names, annotation formats, image conditions, and class distributions make it difficult to combine these datasets or train models that generalize well to real world scenarios. To address these issues, we introduce the GlobalWasteData (GWD) archive, a large scale dataset of 89,807 images across 14 main categories, annotated with 68 distinct subclasses. We compile this novel integrated GWD archive by merging multiple publicly available datasets into a single, unified resource. This GWD archive offers consistent labeling, improved domain diversity, and more balanced class representation, enabling the development of robust and generalizable waste recognition models. Additional preprocessing steps such as quality filtering, duplicate removal, and metadata generation further improve dataset reliability. Overall, this dataset offers a strong foundation for Machine Learning (ML) applications in environmental monitoring, recycling automation, and waste identification, and is publicly available to promote future research and reproducibility.

Figures (6)

• Figure 1: Workflow for creating GWD archive
• Figure 2: Waste classification datasets with their sample sizes
• Figure 3: Geographic distribution of source datasets
• Figure 4: Annotation methods of source datasets
• Figure 5: License types of source datasets