Smart City Gnosys
Smart city article details
| Title | Smartcamairdetect: A Contrastive Approach For Probabilistic Ambient Air Pollution Estimation With Limited Images For Smart City Development |
|---|---|
| ID_Doc | 51817 |
| Authors | Han Y.; Li V.O.; Lam J.C.; Song S.; Mo T. |
| Year | 2025 |
| Published | IEEE Access |
| DOI | http://dx.doi.org/10.1109/ACCESS.2025.3583794 |
| Abstract | Effective air pollution monitoring is crucial for data-driven smart city initiatives. However, the sparse distribution of air quality monitoring stations (AQMSs) often results in localized measurements, which may not accurately represent broader air quality conditions. The cost of extensive AQMS expansion is prohibitive, whereas widely distributed stationary cameras offer a cost-effective alternative by providing high-density imagery for ambient air pollution estimation. Nonetheless, cameras distant from AQMSs or newly installed often lack sufficient labeled data, and conventional models trained on limited datasets tend to overfit, leading to substantial estimation errors. To address these challenges, we introduce SmartCamAirDetect, a novel AI-driven contrastive learning framework to enhance air quality estimation performance, tailored to stationary ground-level camera imagery under sparse labeling conditions. Unlike previous contrastive methods developed for satellite-based data, SmartCamAirDetect is specifically designed to handle the fixed scenery and high temporal variability unique to ground-level images. At the core of SmartCamAirDetect is ContraEst, a contrastive learning model that estimates pollution differentials between images and generates probabilistic pollution assessments by comparing a target image from a limited dataset to multiple images captured by other cameras with ample labels. To refine feature extraction, SmartCamAirDetect synthesizes scenery images to eliminate static background features, enabling the model to focus exclusively on pollution-related information such as sky appearance. SmartCamAirDetect is evaluated across various pollutants using multiple performance metrics. Compared to the baseline deterministic model, ContraEst reduces the mean absolute error by 11.61%, decreases the symmetric mean absolute percentage error by 14.20%, and improves the Pearson correlation coefficient by 50.97%. Incorporating synthetic scenery images further reduces the pinball loss and Winkler score for probabilistic estimations by 5.38% and 5.81%, respectively. These results underscore the efficacy and robustness of SmartCamAirDetect for image-based air pollution estimation, highlighting its potential for fine-grained, AI-driven air quality monitoring in smart cities. In future work, we aim to improve the interpretability and robustness of the scenery synthesis and removal module under dynamic scenes, enhance contrastive pair selection via optimal sampling strategies, incorporate additional probabilistic baselines to better contextualize uncertainty estimation performance, and strengthen generalizability across diverse urban environments. |
| Author Keywords | Air quality; image-based pollution estimation; scenery image synthesis and removal; smart city; smart environment; stationary camera-taken images |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
41912  | 0.899 | Song S.; Li V.O.K.; Lam J.C.K.; Wang Y. | Personalized Ambient Pollution Estimation Based On Stationary-Camera-Taken Images Under Cross-Camera Information Sharing In Smart City | IEEE Internet of Things Journal, 10, 17 (2023) |
| 1343 | 0.877 | Ghose B.; Rehena Z.; Anthopoulos L. | A Deep Learning Based Air Quality Prediction Technique Using Influencing Pollutants Of Neighboring Locations In Smart City | Journal of Universal Computer Science, 28, 8 (2022) |
| 58757 | 0.873 | Njaime M.; Abdallah F.; Snoussi H.; Akl J.; Chaaban K.; Omrani H. | Transfer Learning Based Solution For Air Quality Prediction In Smart Cities Using Multimodal Data | International Journal of Environmental Science and Technology, 22, 3 (2025) |
| 10241 | 0.868 | Xia S.; Xing T.; Chase Q.W.U.; Liu G.; Yang J.; Li K. | Aqmon: A Fine-Grained Air Quality Monitoring System Based On Uav Images For Smart Cities | ACM Transactions on Sensor Networks, 20, 2 (2024) |
| 42245 | 0.866 | Faydi M.; Zrelli A.; Ezzedine T. | Pm2.5 Prediction Using Deep Learning Models | 17th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2023 - Proceedings (2023) |
| 29876 | 0.865 | Bachechi C.; Rollo F.; Po L. | Hypeair: A Novel Framework For Real-Time Low-Cost Sensor Calibration For Air Quality Monitoring In Smart Cities | Ecological Informatics, 81 (2024) |
| 10552 | 0.862 | Neo E.X.; Hasikin K.; Lai K.W.; Mokhtar M.I.; Azizan M.M.; Hizaddin H.F.; Razak S.A.; Yanto | Artificial Intelligence-Assisted Air Quality Monitoring For Smart City Management | PeerJ Computer Science, 9 (2023) |
| 53645 | 0.86 | Essamlali I.; Nhaila H.; El Khaili M. | Supervised Machine Learning Approaches For Predicting Key Pollutants And For The Sustainable Enhancement Of Urban Air Quality: A Systematic Review | Sustainability (Switzerland), 16, 3 (2024) |
| 49184 | 0.859 | Garrido-Hidalgo C.; Solmaz G.; Jacobs T.; Roda-Sanchez L. | Smart Beestricts: Improving The Spatial Resolution Of Air-Quality Data In Madrid Through Transfer Learning | International Journal of Geographical Information Science (2025) |
| 51437 | 0.859 | Chang-Silva R.; Tariq S.; Loy-Benitez J.; Yoo C. | Smart Solutions For Urban Health Risk Assessment: A Pm2.5 Monitoring System Incorporating Spatiotemporal Long-Short Term Graph Convolutional Network | Chemosphere, 335 (2023) |