Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling
Air quality is crucial for both public health and environmental sustainability. An efficient and cost-effective model is essential for accurate air quality predictions and proactive pollution control. However, existing research primarily focuses on single static image analysis, which does not accoun...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-05-01
|
| Series: | Environment International |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0160412025002478 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850279348294123520 |
|---|---|
| author | Maqsood Ahmed Xiang Zhang Yonglin Shen Tanveer Ahmed Shahid Ali Ayaz Ali Aminjon Gulakhmadov Won-Ho Nam Nengcheng Chen |
| author_facet | Maqsood Ahmed Xiang Zhang Yonglin Shen Tanveer Ahmed Shahid Ali Ayaz Ali Aminjon Gulakhmadov Won-Ho Nam Nengcheng Chen |
| author_sort | Maqsood Ahmed |
| collection | DOAJ |
| description | Air quality is crucial for both public health and environmental sustainability. An efficient and cost-effective model is essential for accurate air quality predictions and proactive pollution control. However, existing research primarily focuses on single static image analysis, which does not account for the dynamic and temporal nature of air pollution. Meanwhile, research on video-based air quality estimation remains limited, particularly in achieving accurate multi-pollutant outputs. This study proposes Air Quality Prediction-Mamba (AQP-Mamba), a video-based deep learning model that integrates a structured Selective State Space Model (SSM) with a selective scan mechanism and a hybrid predictor (HP) to estimate air quality. The spatiotemporal forward and backward SSM dynamically adjusts parameters based on input, ensures linear complexity, and effectively captures long-range dependencies by bidirectional processing of spatiotemporal features through four scanning techniques (row-wise, column-wise, and their vertical reversals), which allows the model to accurately track pollutant concentrations and air quality variations over time. Thus, the model efficiently extracts spatiotemporal features from video and simultaneously performs regression (PM2.5, PM10, and AQI), and classification (AQI) tasks, respectively. A high-quality outdoor hourly air quality dataset (LMSAQV) with 13,176 videos collected from six monitoring stations in Lahore, Pakistan, was utilized as the case study. The experimental results demonstrate that the AQP-Mamba significantly outperforms several state-of-the-art models, including VideoSwin-T, VideoMAE, I3D, VTHCL, and TimeSformer. The proposed model achieves strong regression performance (PM2.5: R2 = 0.91, PM10: R2 = 0.90, AQI: R2 = 0.92) and excellent classification metrics: accuracy (94.57 %), precision (93.86 %), recall (94.20 %), and F1-score (93.44 %), respectively. The proposed model delivers consistent, real-time performance with a latency of 1.98 s per video, offering an effective, scalable, and cost-efficient solution for multi-pollutant estimation. This approach has the potential to address gaps in air quality data collected by expensive instruments globally. |
| format | Article |
| id | doaj-art-84417a49cf2a4762aec2ed8a3557f310 |
| institution | OA Journals |
| issn | 0160-4120 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Environment International |
| spelling | doaj-art-84417a49cf2a4762aec2ed8a3557f3102025-08-20T01:49:07ZengElsevierEnvironment International0160-41202025-05-0119910949610.1016/j.envint.2025.109496Low-cost video-based air quality estimation system using structured deep learning with selective state space modelingMaqsood Ahmed0Xiang Zhang1Yonglin Shen2Tanveer Ahmed3Shahid Ali4Ayaz Ali5Aminjon Gulakhmadov6Won-Ho Nam7Nengcheng Chen8National Engineering Research Center of Geographic Information System, School of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaNational Engineering Research Center of Geographic Information System, School of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, China; Corresponding author.National Engineering Research Center of Geographic Information System, School of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaDepartment of Computer Science, School of Engineering and Digital Sciences, Nazarbayev University, Astana, KazakhstanSoftware College, Northeastern University, Shenyang 110169, ChinaDepartment of Cybernetics, Nanotechnology and Data Processing, Faculty of Automatic Control, Electronics and Computer Science, Silesian University of Technology, Akademicka 16, 44-100 Gliwice, PolandResearch Center of Ecology and Environment in Central Asia, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China; Institute of Water Problems, Hydropower and Ecology of the National Academy of Sciences of Tajikistan, Dushanbe 734042, TajikistanSchool of Social Safety and Systems Engineering, Institute of Agricultural Environmental Science, National Agricultural Water Research Center, Hankyong National University, Anseong, Republic of KoreaNational Engineering Research Center of Geographic Information System, School of Geography and Information Engineering, China University of Geosciences, Wuhan 430074, ChinaAir quality is crucial for both public health and environmental sustainability. An efficient and cost-effective model is essential for accurate air quality predictions and proactive pollution control. However, existing research primarily focuses on single static image analysis, which does not account for the dynamic and temporal nature of air pollution. Meanwhile, research on video-based air quality estimation remains limited, particularly in achieving accurate multi-pollutant outputs. This study proposes Air Quality Prediction-Mamba (AQP-Mamba), a video-based deep learning model that integrates a structured Selective State Space Model (SSM) with a selective scan mechanism and a hybrid predictor (HP) to estimate air quality. The spatiotemporal forward and backward SSM dynamically adjusts parameters based on input, ensures linear complexity, and effectively captures long-range dependencies by bidirectional processing of spatiotemporal features through four scanning techniques (row-wise, column-wise, and their vertical reversals), which allows the model to accurately track pollutant concentrations and air quality variations over time. Thus, the model efficiently extracts spatiotemporal features from video and simultaneously performs regression (PM2.5, PM10, and AQI), and classification (AQI) tasks, respectively. A high-quality outdoor hourly air quality dataset (LMSAQV) with 13,176 videos collected from six monitoring stations in Lahore, Pakistan, was utilized as the case study. The experimental results demonstrate that the AQP-Mamba significantly outperforms several state-of-the-art models, including VideoSwin-T, VideoMAE, I3D, VTHCL, and TimeSformer. The proposed model achieves strong regression performance (PM2.5: R2 = 0.91, PM10: R2 = 0.90, AQI: R2 = 0.92) and excellent classification metrics: accuracy (94.57 %), precision (93.86 %), recall (94.20 %), and F1-score (93.44 %), respectively. The proposed model delivers consistent, real-time performance with a latency of 1.98 s per video, offering an effective, scalable, and cost-efficient solution for multi-pollutant estimation. This approach has the potential to address gaps in air quality data collected by expensive instruments globally.http://www.sciencedirect.com/science/article/pii/S0160412025002478PM2.5PM10Air quality index (AQI)Deep learningRegressionClassification |
| spellingShingle | Maqsood Ahmed Xiang Zhang Yonglin Shen Tanveer Ahmed Shahid Ali Ayaz Ali Aminjon Gulakhmadov Won-Ho Nam Nengcheng Chen Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling Environment International PM2.5 PM10 Air quality index (AQI) Deep learning Regression Classification |
| title | Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling |
| title_full | Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling |
| title_fullStr | Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling |
| title_full_unstemmed | Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling |
| title_short | Low-cost video-based air quality estimation system using structured deep learning with selective state space modeling |
| title_sort | low cost video based air quality estimation system using structured deep learning with selective state space modeling |
| topic | PM2.5 PM10 Air quality index (AQI) Deep learning Regression Classification |
| url | http://www.sciencedirect.com/science/article/pii/S0160412025002478 |
| work_keys_str_mv | AT maqsoodahmed lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT xiangzhang lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT yonglinshen lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT tanveerahmed lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT shahidali lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT ayazali lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT aminjongulakhmadov lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT wonhonam lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling AT nengchengchen lowcostvideobasedairqualityestimationsystemusingstructureddeeplearningwithselectivestatespacemodeling |