Visible Waste Index (VWI) — Coded Image Dataset: Three-City Street-Level Imagery Study (Tokyo, Bangkok, Metro Manila)

Source
DataCite
Published
2026
License
Creative Commons Attribution 4.0 International
Themes
Waste Management, Remote Sensing & Satellites, Machine Learning & AI

Get the data from DataCite

About this dataset

This dataset contains adjudicated image-level coding results from a three-city application of the Visible Waste Index (VWI) protocol — a structured coding instrument for measuring neighborhood-scale visible waste from georeferenced street-level imagery (Mapillary and KartaView). The dataset covers Tokyo (23 wards), Bangkok Metropolitan Area, and Metro Manila, comprising 838 image records across 44 grid cells (561 usable images), plus a 64-image intercoder reliability pass independently coded by three coders. Each record includes usability screening, six primary waste component codes (litter_small, dumping_pile, overflow_bin, drain_canal_waste, bulky_waste, open_burning), contextual annotation fields, and derived VWI scores. This dataset accompanies the manuscript: "A Street-Level Imagery Protocol for Measuring Neighborhood Visible Waste: Development, Validation, and Cross-City...

How the data was collected

Each record includes usability screening, six primary waste component codes (litter_small, dumping_pile, overflow_bin, drain_canal_waste, bulky_waste, open_burning), contextual annotation fields, and derived VWI scores.

Browse sustainability datasets on SusanHub