Applying traffic camera and deep learning-based image analysis to predict PM2.5 concentrations

Yanming Liu; Yuxi Zhang; Pei Yu; Tingting Ye; Yiwen Zhang; Rongbin Xu; Shanshan Li; Yuming Guo

doi:10.1016/j.scitotenv.2023.169233

Applying traffic camera and deep learning-based image analysis to predict PM_2.5 concentrations

Sci Total Environ. 2024 Feb 20:912:169233. doi: 10.1016/j.scitotenv.2023.169233. Epub 2023 Dec 12.

Authors

Yanming Liu¹, Yuxi Zhang², Pei Yu¹, Tingting Ye¹, Yiwen Zhang¹, Rongbin Xu¹, Shanshan Li¹, Yuming Guo³

Affiliations

¹ School of Public Health and Preventive Medicine, Monash University, Melbourne, VIC 3004, Australia.
² School of Public Health and Preventive Medicine, Monash University, Melbourne, VIC 3004, Australia; School of Life and Environmental Sciences, the University of Sydney, Sydney, NSW 2006, Australia.
³ School of Public Health and Preventive Medicine, Monash University, Melbourne, VIC 3004, Australia. Electronic address: YumingGuo@monash.edu.

PMID: 38097087
DOI: 10.1016/j.scitotenv.2023.169233

Abstract

Background: Air pollution has caused a significant burden in terms of mortality and mobility worldwide. However, the current coverage of air quality monitoring networks is still limited.

Objective: This study aims to apply a novel approach to convert the existing traffic cameras into sensors measuring particulate matter with a diameter of 2.5 μm or less (PM_2.5) so that the coverage of PM_2.5 monitoring could be expanded without extra cost.

Methods: In our study, the traffic camera images were collected at a rate of 4 images/h and the corresponding hourly PM_2.5 concentration was collected from the reference grade PM_2.5 station 3 km away. A customized neural network model was trained to obtain the PM_2.5 concentration from images followed by a random forest model to predict the hourly PM_2.5 concentration. The saliency maps and the feature importance were utilized to interpret the neural network.

Results: Proposed novel approach has a high prediction performance to predict hourly PM_2.5 from traffic camera images, with a root mean square error (RMSE) of 0.76 μg/m³ and a coefficient of determination (R²) of 0.98. The saliency map shows neural network focuses on unobstructed far-end road surfaces while the random forest feature importance highlights the first quarter image's significance. The model performance is robust whether weather conditions are controlled or not.

Conclusion: Our study provided a practical approach to converting the existing traffic cameras into PM_2.5 sensors. The deep learning method based on the Resnet architecture in our study can broaden the coverage of PM_2.5 monitoring with no additional infrastructure needed.

Keywords: Air quality; Deep learning; Image analysis; Machine learning; PM(2.5); Traffic camera.