Beyond here and now : Evaluating pollution estimation across space and time from street view images with deep learning
Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved..
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.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:903 |
|---|---|
| Enthalten in: |
The Science of the total environment - 903(2023) vom: 10. Dez., Seite 166168 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Nathvani, Ricky [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Air pollution |
|---|
| Anmerkungen: |
Date Revised 06.03.2024 published: Print-Electronic Citation Status PubMed-not-MEDLINE |
|---|
| doi: |
10.1016/j.scitotenv.2023.166168 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM360857345 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM360857345 | ||
| 003 | DE-627 | ||
| 005 | 20240306232352.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.scitotenv.2023.166168 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1318.xml |
| 035 | |a (DE-627)NLM360857345 | ||
| 035 | |a (NLM)37586538 | ||
| 035 | |a (PII)S0048-9697(23)04793-9 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Nathvani, Ricky |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Beyond here and now |b Evaluating pollution estimation across space and time from street view images with deep learning |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Revised 06.03.2024 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status PubMed-not-MEDLINE | ||
| 520 | |a Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved. | ||
| 520 | |a 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 | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Air pollution | |
| 650 | 4 | |a Computer vision | |
| 650 | 4 | |a Deep learning | |
| 650 | 4 | |a Environmental modelling | |
| 650 | 4 | |a Noise pollution | |
| 650 | 4 | |a Street-view images | |
| 700 | 1 | |a D, Vishwanath |e verfasserin |4 aut | |
| 700 | 1 | |a Clark, Sierra N |e verfasserin |4 aut | |
| 700 | 1 | |a Alli, Abosede S |e verfasserin |4 aut | |
| 700 | 1 | |a Muller, Emily |e verfasserin |4 aut | |
| 700 | 1 | |a Coste, Henri |e verfasserin |4 aut | |
| 700 | 1 | |a Bennett, James E |e verfasserin |4 aut | |
| 700 | 1 | |a Nimo, James |e verfasserin |4 aut | |
| 700 | 1 | |a Moses, Josephine Bedford |e verfasserin |4 aut | |
| 700 | 1 | |a Baah, Solomon |e verfasserin |4 aut | |
| 700 | 1 | |a Hughes, Allison |e verfasserin |4 aut | |
| 700 | 1 | |a Suel, Esra |e verfasserin |4 aut | |
| 700 | 1 | |a Metzler, Antje Barbara |e verfasserin |4 aut | |
| 700 | 1 | |a Rashid, Theo |e verfasserin |4 aut | |
| 700 | 1 | |a Brauer, Michael |e verfasserin |4 aut | |
| 700 | 1 | |a Baumgartner, Jill |e verfasserin |4 aut | |
| 700 | 1 | |a Owusu, George |e verfasserin |4 aut | |
| 700 | 1 | |a Agyei-Mensah, Samuel |e verfasserin |4 aut | |
| 700 | 1 | |a Arku, Raphael E |e verfasserin |4 aut | |
| 700 | 1 | |a Ezzati, Majid |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t The Science of the total environment |d 1972 |g 903(2023) vom: 10. Dez., Seite 166168 |w (DE-627)NLM000215562 |x 1879-1026 |7 nnns |
| 773 | 1 | 8 | |g volume:903 |g year:2023 |g day:10 |g month:12 |g pages:166168 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.scitotenv.2023.166168 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 903 |j 2023 |b 10 |c 12 |h 166168 | ||