MDWD: A Street-Level Dataset for Municipal Solid Waste Detection in Dense Urban Environments

2026-07-31

AI Summary

This study addresses the absence of benchmark datasets integrating street-view imagery with instance-level localization and classification in municipal solid waste management research. We present the first urban-scene-oriented benchmark dataset for solid waste detection, comprising 3,697 high-resolution street-view images and 11,461 annotated instances across five categories of household waste in Malta, capturing diverse urban environmental variations. Leveraging this dataset, we conduct a cross-model evaluation of YOLO variants and Transformer-based architectures, including RF-DETR-M. Experimental results demonstrate that RF-DETR-M achieves 94.49% mAP50 and 93.56% F1-score on the test set, while lightweight models also maintain high accuracy, confirming the dataset’s robust support for diverse object detection frameworks.

Abstract

Automated visual monitoring of urban environments is a growing Computer Vision research area, but municipal solid waste detection remains under-represented in dedicated benchmark resources. Existing waste-related datasets predominantly address individual litter detection, aerial imagery, or image-level classification, and none simultaneously provide street-level imagery, instance-level localization, and categorization of domestic waste streams within a structured municipal collection context. This paper introduces the Maltese Domestic Waste Dataset (MDWD), a street-level benchmark comprising 3,697 high-resolution images and 11,461 manually annotated instances across five domestic waste categories representative of Malta's municipal collection system. The dataset captures substantial variation in location, illumination, object scale, occlusion, and urban context. To establish reproducible baselines, a cross-architecture benchmark is conducted across multiple generations of the YOLO family and a transformer-based detector. On the test set, RF-DETR-M achieves the strongest overall performance with an mAP50 of 94.49% and an F1-score of 93.56%, whilst smaller-capacity variants maintain competitive accuracy at substantially reduced parameter counts. These results indicate that MDWD supports effective training across both compact real-time detectors and transformer-based models, establishing a benchmark for future research in vision-based municipal waste monitoring.

Problem

• municipal solid waste detection
• street-level dataset
• instance-level localization
• domestic waste categorization
• urban visual monitoring

Innovation

• street-level waste detection
• instance-level annotation
• municipal solid waste dataset
• YOLO benchmarking
• vision-based urban monitoring