• Home
  • Search
  • A Deep Learning Approach for Meter-Scale Air Quality Estimation in Urban Environments Using Very High-Resolution Satellite Imagery
  • https://doi.org/10.1289/isee.2021.o-to-042Copy DOI Icon

A Deep Learning Approach for Meter-Scale Air Quality Estimation in Urban Environments Using Very High-Resolution Satellite Imagery

Show More
  • Abstract
  • Literature Map
  • Similar Papers
Abstract

BACKGROUND AND AIM: Air Quality (AQ) is currently measured using ground monitoring stations which require funding and infrastructure to support and lack spatial coverage. In contrast, high-resolution commercial satellite imagery (⪯2 m/pixel) can be produced for almost any location on Earth and is readily available. The goal of this project is to develop a pipeline that uses MAXAR satellite imagery to produce meter-scale, continuous maps of AQ for any city around the globe, with greater resolution and coverage than existing methods. METHODS: We developed a Deep Neural Network (DNN) model that can produce AQ estimates for previously unseen urban locations using satellite imagery. The model is based on the VGG-16 pretrained neural network architecture. We used this architecture to extract spatial features from satellite imagery that are passed on to Fully Connected layers to produce an estimate of PM2.5 and NO2 concentrations. We partitioned satellite imagery into 612,248 patches and fed these patches into the model to produce a 100m (200m) continuous grid of PM2.5 (NO2) estimates. RESULTS:We fed the high-resolution (0.5m–2.5m) commercial satellite imagery to a DNN to learn the association between visual urban features and air pollutants. The developed model, which solely uses satellite imagery, was tested and evaluated using both ground monitoring observations and land-use regression modeled PM2.5 and NO2 concentrations over London, Vancouver (BC), Los Angeles, and New York City. The results demonstrate a low error with a total RMSE 2 µg/m3 and highlight the contribution of specific urban features, such as green areas and roads, to continuous meter-scale AQ estimations. CONCLUSIONS:This approach offers promise for scaling to global applications in developed and developing urban environments. Further analysis on domain transferability will enable application of a parsimonious model based merely on satellite images to create meter-scale AQ maps in developing cities, where current and historical ground data is limited. KEYWORDS: Air quality. Remote Sensing, Urban environment, Deep learning, Satellite Imagery

Similar Papers
  • Research Article
  • Citations31

Oil palm trees detection and counting on Microsoft Bing Maps Very High Resolution (VHR) satellite imagery and Unmanned Aerial Vehicles (UAV) data using image processing thresholding approach

  • Oct 27, 2022
  • Ecological Informatics
  • Yoga Cahya Putra +2
  • Conference Article
  • Citations1

High Resolution Satellite Imagery: From Spies to Pipeline Management

  • Oct 01, 2000
  • Steve Adam +1
  • PDF
  • Research Article
  • Citations3

Eutrophic Status Assessment Based on Very High-Resolution Satellite Imagery in the Coastline Environment of Korea

  • Jan 16, 2023
  • Pollutants
  • Muhammad Mazhar Iqbal +6
  • PDF
  • Research Article
  • Citations42

Exploring the Potential of High-Resolution Satellite Imagery for the Detection of Soybean Sudden Death Syndrome

  • Apr 09, 2020
  • Remote Sensing
  • Muhammad M Raza +3
  • Dissertation
  • Citations5

Wavelet Analysis and Classification of Urban Environment Using High -Resolution Multispectral Image Data.

  • Jan 01, 2001
  • Soe Myint
  • Research Article
  • Citations11

Archaeological documentation of wood caribou fences using unmanned aerial vehicle and very high-resolution satellite imagery in the Mackenzie Mountains, Northwest Territories

  • Sep 01, 2020
  • Journal of Unmanned Vehicle Systems
  • Jurjen Van Der Sluijs +4
  • Conference Article
  • Citations5

Multipath sparse coding for scene classification in very high resolution satellite imagery

  • Oct 15, 2015
  • Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE
  • Jiayuan Fan +2
  • PDF
  • Research Article
  • Citations12

Estimation of total population size of southern elephant seals (Mirounga leonina) on Kerguelen and Crozet Archipelagos using very high-resolution satellite imagery

  • Jun 12, 2023
  • Frontiers in Marine Science
  • Joris Laborie +5
  • Research Article
  • Citations23

A new method for road detection in urban areas using high-resolution satellite images and Lidar data based on fuzzy nearest-neighbor classification and optimal features

  • Apr 26, 2016
  • Arabian Journal of Geosciences
  • Asghar Milan Lak +2
  • Research Article

High-Resolution Remote Sensing and People-to-Pixel Integration for Mapping Farmland Abandonment in Central Himalayan Villages

  • Nov 15, 2025
  • Remote Sensing
  • Basanta Paudel +5
  • Supplementary Content
  • Citations104

Shadow information recovery in urban areas from very high resolution satellite imagery

  • Jul 31, 2007
  • International Journal of Remote Sensing
  • Y Chen +3
  • Research Article
  • Citations189

An automatic method for road extraction in rural and semi-urban areas starting from high resolution satellite imagery

  • Dec 10, 2004
  • Pattern Recognition Letters
  • J.B Mena +1
  • PDF
  • Research Article
  • Citations36

Detection of Tailings Dams Using High-Resolution Satellite Imagery and a Single Shot Multibox Detector in the Jing–Jin–Ji Region, China

  • Aug 14, 2020
  • Remote Sensing
  • Qingting Li +6
  • Research Article
  • Citations13

Monitoring coastal water flow dynamics using sub-daily high-resolution SkySat satellite and UAV-based imagery

  • May 02, 2022
  • Water Research
  • Kasper Johansen +4
  • Research Article
  • Citations16

Site-Specific Management of Cotton Root Rot Using Airborne and High-Resolution Satellite Imagery and Variable-Rate Technology

  • Jan 01, 2018
  • Transactions of the ASABE
  • Chenghai Yang +6
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.