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

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) so that the coverage of PM 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 concentration was collected from the reference grade PM station 3 km away. A customized neural network model was trained to obtain the PM concentration from images followed by a random forest model to predict the hourly PM 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 from traffic camera images, with a root mean square error (RMSE) of 0.76 μg/m and a coefficient of determination (R) 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 sensors. The deep learning method based on the Resnet architecture in our study can broaden the coverage of PM monitoring with no additional infrastructure needed.