A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques
Particulate pollution (PM2.5) is an important concern in Asian countries owing to its health hazards. When planning outdoor activities, understanding the PM2.5 concentration measurement is essential. Because of the lower number of government-run Air Quality Monitoring Stations, other options for obt...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10813353/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583210420142080 |
|---|---|
| author | Anupam Kamble Somrawee Aramkul Paskorn Champrasert |
| author_facet | Anupam Kamble Somrawee Aramkul Paskorn Champrasert |
| author_sort | Anupam Kamble |
| collection | DOAJ |
| description | Particulate pollution (PM2.5) is an important concern in Asian countries owing to its health hazards. When planning outdoor activities, understanding the PM2.5 concentration measurement is essential. Because of the lower number of government-run Air Quality Monitoring Stations, other options for obtaining location-specific PM2.5 concentration values are sought. This paper proposes using photo image processing to estimate the PM2.5 concentration. This research aims to improve the efficacy and reduce the computational complexity of the PM2.5 concentration estimation process. The proposed Efficient PM2.5 estimation framework uses EfficientNet-B1 and BiLSTM to estimate PM2.5 concentrations. Met-EfficientNet-B1-BiLSTM was designed and implemented to incorporate the meteorological features - temperature, wind speed, and humidity to further improve the estimation accuracy. The EfficientNet-B1 neural network is applied in the image feature vector extraction process. EfficientNet-B1, with a resolution of <inline-formula> <tex-math notation="LaTeX">$240\times 240$ </tex-math></inline-formula> pixels, was determined to be the optimal variant of EfficientNet for a small dataset of images needed for the estimation of the PM2.5 concentration value. The BiLSTM was used for the regression of these image features with PM2.5 concentration values to obtain the estimated PM2.5 concentration. A dataset comprising HDR and non-HDR images was explicitly created for this study to compare the types of images that improve the accuracy of PM2.5 concentration estimation and the feature extraction process. The proposed Efficient PM2.5 estimation framework reduces computational complexity and outperforms ResNet-18-LSTM by improving the efficacy by 5.75% in MAE and 11.43% in SMAPE metrics. The proposed Efficient PM2.5 estimation framework demonstrates that the mobile image can be efficiently used for PM2.5 concentration estimation. |
| format | Article |
| id | doaj-art-19aa32121c824b71873d81de70440592 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-19aa32121c824b71873d81de704405922025-01-29T00:01:18ZengIEEEIEEE Access2169-35362025-01-0113161961620710.1109/ACCESS.2024.352196610813353A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning TechniquesAnupam Kamble0https://orcid.org/0009-0009-0025-2498Somrawee Aramkul1https://orcid.org/0000-0003-1936-9054Paskorn Champrasert2https://orcid.org/0000-0002-1882-2569Data Science Consortium, Faculty of Engineering, Chiang Mai University, Chiang Mai, ThailandDepartment of Computer, Faculty of Science and Technology, Chiang Mai Rajabhat University, Chiang Mai, ThailandOASYS Research Group, Faculty of Engineering, Chiang Mai University, Chiang Mai, ThailandParticulate pollution (PM2.5) is an important concern in Asian countries owing to its health hazards. When planning outdoor activities, understanding the PM2.5 concentration measurement is essential. Because of the lower number of government-run Air Quality Monitoring Stations, other options for obtaining location-specific PM2.5 concentration values are sought. This paper proposes using photo image processing to estimate the PM2.5 concentration. This research aims to improve the efficacy and reduce the computational complexity of the PM2.5 concentration estimation process. The proposed Efficient PM2.5 estimation framework uses EfficientNet-B1 and BiLSTM to estimate PM2.5 concentrations. Met-EfficientNet-B1-BiLSTM was designed and implemented to incorporate the meteorological features - temperature, wind speed, and humidity to further improve the estimation accuracy. The EfficientNet-B1 neural network is applied in the image feature vector extraction process. EfficientNet-B1, with a resolution of <inline-formula> <tex-math notation="LaTeX">$240\times 240$ </tex-math></inline-formula> pixels, was determined to be the optimal variant of EfficientNet for a small dataset of images needed for the estimation of the PM2.5 concentration value. The BiLSTM was used for the regression of these image features with PM2.5 concentration values to obtain the estimated PM2.5 concentration. A dataset comprising HDR and non-HDR images was explicitly created for this study to compare the types of images that improve the accuracy of PM2.5 concentration estimation and the feature extraction process. The proposed Efficient PM2.5 estimation framework reduces computational complexity and outperforms ResNet-18-LSTM by improving the efficacy by 5.75% in MAE and 11.43% in SMAPE metrics. The proposed Efficient PM2.5 estimation framework demonstrates that the mobile image can be efficiently used for PM2.5 concentration estimation.https://ieeexplore.ieee.org/document/10813353/PM2.5CNNsEfficientNetResNetMBConvRNNs |
| spellingShingle | Anupam Kamble Somrawee Aramkul Paskorn Champrasert A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques IEEE Access PM2.5 CNNs EfficientNet ResNet MBConv RNNs |
| title | A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques |
| title_full | A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques |
| title_fullStr | A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques |
| title_full_unstemmed | A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques |
| title_short | A Mobile Image-Driven PM2.5 Estimation Framework Using Deep Learning Techniques |
| title_sort | mobile image driven pm2 5 estimation framework using deep learning techniques |
| topic | PM2.5 CNNs EfficientNet ResNet MBConv RNNs |
| url | https://ieeexplore.ieee.org/document/10813353/ |
| work_keys_str_mv | AT anupamkamble amobileimagedrivenpm25estimationframeworkusingdeeplearningtechniques AT somraweearamkul amobileimagedrivenpm25estimationframeworkusingdeeplearningtechniques AT paskornchamprasert amobileimagedrivenpm25estimationframeworkusingdeeplearningtechniques AT anupamkamble mobileimagedrivenpm25estimationframeworkusingdeeplearningtechniques AT somraweearamkul mobileimagedrivenpm25estimationframeworkusingdeeplearningtechniques AT paskornchamprasert mobileimagedrivenpm25estimationframeworkusingdeeplearningtechniques |