AQPDCITY Dataset: Picture-Based PM Monitoring in the Urban Area of Big Cities

TL;DR

This paper introduces the AQPDCITY dataset, a large collection of city photos for PM air quality monitoring, highlighting the limitations of existing sensor-based methods and evaluating state-of-the-art image-based approaches.

Contribution

The creation of a new large-scale city photo dataset for particulate matter monitoring and the assessment of current image-based methods' performance on it.

Findings

Existing methods perform poorly on the AQPDCITY dataset.

The dataset provides a new benchmark for image-based air quality monitoring.

Sensor-based methods have limitations in dense urban environments.

Abstract

Since Particulate Matters (PMs) are closely related to people's living and health, it has become one of the most important indicator of air quality monitoring around the world. But the existing sensor-based methods for PM monitoring have remarkable disadvantages, such as low-density monitoring stations and high-requirement monitoring conditions. It is highly desired to devise a method that can obtain the PM concentration at any location for the following air quality control in time. The prior works indicate that the PM concentration can be monitored by using ubiquitous photos. To further investigate such issue, we gathered 1,500 photos in big cities to establish a new AQPDCITY dataset. Experiments conducted to check nine state-of-the-art methods on this dataset show that the performance of those above methods perform poorly in the AQPDCITY dataset.