Abstract:
Advances in computer vision, driven by deep learning, allows for the inference of environmental pollution and its
potential sources from images. The spatial and temporal generalisability of image-based pollution models is
crucial in their real-world application, but is currently understudied, particularly in low-income countries where infrastructure for measuring the complex patterns of pollution is limited and modelling may therefore provide
the most utility. We employed convolutional neural networks (CNNs) for two complementary classification
models, in both an end-to-end approach and as an interpretable feature extractor (object detection), to estimate
spatially and temporally resolved fine particulate matter (PM2.5) and noise levels in Accra, Ghana. Data used for
training the models were from a unique dataset of over 1.6 million images collected over 15 months at 145
representative locations across the city, paired with air and noise measurements. Both end-to-end CNN and
object-based approaches surpassed null model benchmarks for predicting PM2.5 and noise at single locations, but
performance deteriorated when applied to other locations. Model accuracy diminished when tested on images
from locations unseen during training, but improved by sampling a greater number of locations during model
training, even if the total quantity of data was reduced. The end-to-end models used characteristics of images
associated with atmospheric visibility for predicting PM2.5, and specific objects such as vehicles and people for
noise. The results demonstrate the potential and challenges of image-based, spatiotemporal air pollution and
noise estimation, and that robust, environmental modelling with images requires integration with traditional
sensor networks.
