IJEP 44(15): 1318-1326 : Vol. 44 Issue. 15 (Conference 2024)
1. D.Y. Patil University, Vijay Patil School of Management, Department of Business and Management, Navi Mumbai – 400 706, Maharashtra, India
2. Dr. Vishwanth Karad MIT World Peace University, School of Business, Department of Business and Management, Pune – 411 038, Maharashtra, India
3. Bharati Vidyapeeth (Deemed to be University), Institute of Management and Rural Development Administration, Department of Management, Sangili – 416 416, Maharashtra, India
4. Concordia University of Edmonton, Faculty of Science, Department of Science, Edmonton, Canada
5. Royal Global University, Royal School of Commerce, Guwahati – 781 035, Assam, India
Abstract
Air pollution poses significant health and environmental challenges in today’s urban centres, necessitating advanced detection and control strategies. In this study, we focus on leveraging image data for air pollution identification in smart cities. The air quality image dataset used in our research is sourced from Kaggle, titled ‘air pollution image dataset from India and Nepal’. The dataset comprises 12,240 images, each with dimensions of 224 x 224 pixels, categorized into distinct air quality index (AQI) levels, ranging from good to hazardous/severe. To enhance the quality and suitability of the image data for analysis, we perform preprocessing steps, including grayscale conversion, noise removal and rescaling. Subsequently, we extract essential features from the images using the gray-level co-occurrence matrix (GLCM) method, a well-established approach for texture classification in image analysis. Furthermore, to optimize extracted features, we employ two powerful optimization algorithms: elephant search algorithm (ESA) and cat swarm optimization (CSO). The GLCM features, as well as the optimized ESA and CSO features, are fed into a convolutional neural network (CNN) model for classification. The performance of the CNN model is thoroughly evaluated and most effective optimization technique is identified. The chosen optimization technique is then implemented in a mobile app tailored for smart cities. The mobile app allows individuals to monitor their surrounding environment’s air quality by capturing and uploading images from their surroundings. The app processes the uploaded images and displays the corresponding air pollution level based on the AQI categories. In the event of the air pollution level being deemed ‘unhealthy’ or ‘hazardous,’ the app accesses the user’s location data and automatically notifies the local government. This real-time information empowers the government to take prompt and appropriate actions to mitigate pollution sources and improve air quality in the affected areas. The deployment of this mobile app in smart cities fosters a citizen-centric approach to air pollution control. The integration of artificial intelligent (AI) solutions for air pollution detection, optimization and mobile app deployment presents a promising step toward building smart and environmentally conscious cities for a better future.
Keywords
Air pollution, Images, Kaggle, Pre-process, Feature extraction, Artificial intelligence, Mobile app
References
- Manisalidis, I., et al. 2020. Environmental and health impacts of air pollution: A review. Frontiers Public Health. 14.
- Choudhary, M.P. and V. Garg. 2013. Causes, consequences and control of air pollution. Conference on All India seminar on methodologies for air pollution control. Jaipur, Rajasthan.
- Ko, F.W.S. and D.S.C. Hui. 2012. Air pollution and chronic obstructive pulmonary disease. Respirol., 17(3): 395-401.
- Holland, M.R. 1995. Assessment of the economic costs of damage caused by air pollution. Water Air Soil Poll., 85: 2583-2588.
- Welburn, A. 1988. Air pollution and acid rain: The biological impact. Longman Higher Education.
- Mukundan, A., et al. 2022. Air pollution detection using a novel snap-shot hyperspectol imaging technique. Sensors. 22(16): 6231.
- Kow, P.Y., et al. 2022. Real-time image-based air quality estimation by deep learning neural networks. J. Env. Manage., 307: 114560.
- Chu, Y., et al. 2023. Detection of air pollution in urban areas using monitoring images. Atmos., 14 (5): 772.
- Wang, J., et al. 2022. An air quality index prediction model based on CNN-ILSTM. Sci. Reports. 12(1): 8373.
- Wong, C.J., et al. 2009. Development of low-cost air quality monitoring system by using image processing technique. In Image processing. Ed Y.S. Chen. DOI: 10.5772/7048.
- Sulistiyanti, S.R., F.X. Setyawan and M. Komarudin. 2019. Detection air pollution based on infrared image processing. Telkomnika. 17(4): 1796-1802.
- Zhu, Z., et al. 2019. Image and spectrum based deep feature analysis for particle matter estimation with weather information. IEEE International Conference on Image processing (ICIP). Proceedings, pp 3427-3431.
- Chaki, J. and N. Dey. 2020. Texture feature extraction techniques for image recognition. Springer, Singapore.
- Hall-Beyer, M. 2000. GLCM texture: A tutorial v. 3.0. National Council on Geographic Information and Analysis Remote Sensing core curriculum. 3(1): 75.
- Mohanaiah, P., P. Sathyanarayana and L. Guru-kumar. 2013. Image texture feature extraction using GLCM approach. Int. J. Sci. Res., 3(5):
1-5. - Deb, S., S. Fong and Z. Tian. 2015. Elephant search algorithm for optimization problems. Tenth International Conference on Digitial information management (ICDIM). IEEE. Proceedings, pp 249-255.
- Deb, S., et al. 2016. Finding approximate solutions of NP-hard optimization and TSP problems using elephant search algorithm. J. Supercomputing. 72: 3960-3992.
- Chu, S.C. and P.W. Tsai. 2007. Computational intelligence based on the behaviour of cats. Int. J. Innov. Comput. Inf. Cont., 3(1): 163-173.
- Lin, S.C. and H.Y. Chien. 2009. CSO-based feature selection and parameter optimization for support vector machine. 2009 Joint Conferences on pervasive computing (JCPC). Proceedings, pp 783-788.
- Ahmed, A.M., T.A. Rashid and S.M. Saeed. 2020. Cat swarm optimization algorithm: A survey and performance evaluation. Hindawi Computational Intelligence Neurosci., 2020: 4854895.
- Albawi, S., T.A. Mohammed and S. Al-Zawi. 2017. Understanding of a convolutional neural network. International Conference on Engineering and Technology (ICET). IEEE. Proceedings, pp 1-6.
- Aghdam, H.H. and E.J. Heravi. 2014. Guide to convolutional neural networks: A practical application to traffic-sign detection and classification. Springer Cham.
- Yang, B., et al. 2015. Convolutional channel features. International Conference on Computer vision. Proceedings, pp 82-90.
- Bannerjee, C., T. Mukherjee and E. Pasiliao Jr. 2019. An empirical study on generalization of the ReLu activation function. 2019 ACM Southeast Conference. Proceedings, pp 164-167.
- Jeczmionek, E. and P.A. Kowalski. 2021. Flattening layer pruning in convolutional neural networks. Symmetry. 13(7): 1147.
