Assessing Smoke Particle Contamination on Firefighter Personal Protective Equipment using Generative AI and Edge Detection Models
DOI:
https://doi.org/10.63456/tsrj-2-3-56Keywords:
Smoke analysis, Wildland-oilfield interface fire, Generative Artificial Intelligence, Image edge detection, Occupational safety, Personal Protective EquipmentAbstract
Firefighters at a wildland-oilfield interface (WOI) encounter complex smoke particle mixtures from biomass and hydrocarbons. These particulates settle onto Personal Protective Equipment (PPE) surfaces, remaining long after fire suppression activities. These attached smoke particles create a severe occupational hazard through secondary exposure during movement and the donning/doffing of contaminated gear. Despite the occupational risk, various commercial particle detection and size distribution techniques are destructive or requires liquid extraction of smoke particles for analysis. A non-destructive technique facilitates the understanding of particle size, and distribution onto the fabric surface. This study provides a comparative analysis of particle detection methods using non-destructive, multiple imaging technique through scanning electron microscopy. Following a comparison of traditional edge detection algorithms (Canny, LoG and Otsu) with Generative AI, An optimized detection method is established across multiple PPE textile substrates, to ensure broad applicability of the optimised detection technique. The technique successfully identifies and measures respirable (≤ 2.5 μm) and ultrafine (≤1 μm) particles on textile surfaces. Analysis of wood and oil smoke revealed high concentrations of ultrafine particles, highlighting the invisible threat they pose. Ultimately, this research aims to establish a universally applicable detection framework for smoke particles deposited onto PPE surfaces.
References
[1] Adetona, O., Reinhardt, T. E., Domitrovich, J., Broyles, G., Adetona, A. M., Kleinman, M. T., Ottmar, R. D., & Naeher, L. P. (2016). Review of the health effects of wildland fire smoke on wildland firefighters and the public. Inhalation Toxicology, 28(3), 95–139. https://doi.org/10.3109/08958378.2016.1145771
[2] Bates, M. N. (2007). Registry-based case–control study of cancer in California firefighters. American Journal of Industrial Medicine, 50(5), 339–344. https://doi.org/10.1002/ajim.20446
[3] Baxter, C. S., Hoffman, J. D., Knipp, M. J., Reponen, T., & Haynes, E. N. (2014). Exposure of Firefighters to Particulates and Polycyclic Aromatic Hydrocarbons. Journal of Occupational and Environmental Hygiene, 11(7), D85–D91. https://doi.org/10.1080/15459624.2014.890286
[4] Bolya, D., Foley, S., Hays, J., & Hoffman, J. (2020). TIDE: A General Toolbox for Identifying Object Detection Errors. In A. Vedaldi, H. Bischof, T. Brox, & J.-M. Frahm (Eds.), Computer Vision – ECCV 2020 (Vol. 12348, pp. 558–573). Springer International Publishing. https://doi.org/10.1007/978-3-030-58580-8_33
[5] Classification: Accuracy, recall, precision, and related metrics | Machine Learning | Google for Developers. (2026, January 12). https://developers.google.com/machine-learning/crash-course/classification/accuracy-precision-recall
[6] Datta, M., Nandi, S., Nath, D., Debnath, S., & Basu, G. (2022). Fiber Cross Sectional Shape: A Persuadable Feature for Assessment of Acoustic Noise Control Property of Some Natural Fibers and Its Geometrical Modeling. Journal of Natural Fibers, 19(17), 15600–15613. https://doi.org/10.1080/15440478.2022.2131680
[7] Davydiuk, N., Krieg, E., Gaitzsch, J., McCall, P. M., Auernhammer, G. K., Yang, M., Tracy, J. B., Bals, S., Parak, W. J., Kotov, N. A., Liz-Marzán, L. M., Fery, A., Faria, M., & Besford, Q. A. (2025). The rising danger of AI-generated images in nanomaterials science and what we can do about it. Nature Nanotechnology, 20(9), 1174–1177. https://doi.org/10.1038/s41565-025-02009-9
[8] Demers, P., Martinsen, J. I., Weiderpass, E., Kjaerheim, K., Lynge, E., Sparen, P., & Pukkala, E. (2011). Cancer incidence among Nordic firefighters. Occupational and Environmental Medicine, 68(Suppl_1), A19–A20. https://doi.org/10.1136/oemed-2011-100382.60
[9] Dilger, M., Armant, O., Ramme, L., Mülhopt, S., Sapcariu, S. C., Schlager, C., Dilger, E., Reda, A., Orasche, J., Schnelle-Kreis, J., Conlon, T. M., Yildirim, A. Ö., Hartwig, A., Zimmermann, R., Hiller, K., Diabaté, S., Paur, H.-R., & Weiss, C. (2023). Systems toxicology of complex wood combustion aerosol reveals gaseous carbonyl compounds as critical constituents. Environment International, 179, 108169. https://doi.org/10.1016/j.envint.2023.108169
[10] Dilger, M., Orasche, J., Zimmermann, R., Paur, H.-R., Diabaté, S., & Weiss, C. (2016). Toxicity of wood smoke particles in human A549 lung epithelial cells: The role of PAHs, soot and zinc. Archives of Toxicology, 90(12), 3029–3044. https://doi.org/10.1007/s00204-016-1659-1
[11] Du, M., Wang, Y., Xie, L., Deng, G., Guo, J., Han, B., Chen, Z.-H., Rui, C., Han, J., Chen, Y., Zhao, Y., Cao, R., Wang, F., Li, K., Wang, Y., He, Y., & Feng, X. (2026). A foundation AI model enhances electron microscopy image analysis. bioRxiv, 2026.02.28.708664. https://doi.org/10.64898/2026.02.28.708664
[12] Horn, G. P., Dow, N. W., & Neumann, D. L. (2024). Pilot Study on Fire Effluent Condensate from Full Scale Residential Fires. Fire Technology, 60(1), 1–18. https://doi.org/10.1007/s10694-023-01487-4
[13] Hosseini, S., Qi, L., Cocker, D., Weise, D., Miller, A., Shrivastava, M., Miller, J. W., Mahalingam, S., Princevac, M., & Jung, H. (2010). Particle size distributions from laboratory-scale biomass fires using fast response instruments. Atmos. Chem. Phys. 10: 8065-8076. https://doi.org/10.5194/acp-10-8065-2010
[14] Hurley, M. J., Gottuk, D., Hall, J. R., Harada, K., Kuligowski, E., Puchovsky, M., Torero, J., Watts, J. M., & Wieczorek, C. (Eds.). (2016). SFPE Handbook of Fire Protection Engineering. Springer New York. https://doi.org/10.1007/978-1-4939-2565-0
[15] Jerez, D., Stuart, E., Schmitt, K., Guerrero-Given, D., Christie, J. M., Hahn, W. E., Kamasawa, N., & Smirnov, M. S. (2021). A deep learning approach to identifying immunogold particles in electron microscopy images. Scientific Reports, 11(1), 7771. https://doi.org/10.1038/s41598-021-87015-2
[16] Kamano, K., Jaroenkhasemmeesuk, C., Chaisartra, C., Thoopkaew, T., & Tippayawong, N. (2022). Glass separation process for recycling of solar photovoltaic panels by microwave heating. 020002. https://doi.org/10.1063/5.0115199
[17] Laing, J. R., Jaffe, D. A., & Hee, J. R. (2016). Physical and optical properties of aged biomass burning aerosol from wildfires in Siberia and the Western USA at the Mt. Bachelor Observatory. Atmospheric Chemistry and Physics, 16(23), 15185–15197. https://doi.org/10.5194/acp-16-15185-2016
[18] LeMasters, G. K., Genaidy, A. M., Succop, P., Deddens, J., Sobeih, T., Barriera-Viruet, H., Dunning, K., & Lockey, J. (2006). Cancer Risk Among Firefighters: A Review and Meta-analysis of 32 Studies: Journal of Occupational and Environmental Medicine, 48(11), 1189–1202. https://doi.org/10.1097/01.jom.0000246229.68697.90
[19] Li, D., Pang, P. C.-I., & Lam, C.-K. (2025). Contrast-Invariant Edge Detection: A Methodological Advance in Medical Image Analysis. Applied Sciences, 15(2), 963. https://doi.org/10.3390/app15020963
[20] Lindeberg, T. (1994). Scale-Space Theory in Computer Vision. Springer US. https://doi.org/10.1007/978-1-4757-6465-9
[21] Lu, S., Bhattarai, C., Samburova, V., & Khlystov, A. (2025). Particle size distributions of wildfire aerosols in the western USA. Environmental Science: Atmospheres, 5(4), 502–516. https://doi.org/10.1039/D5EA00007F
[22] Marr, D., & Hildreth, E. (1980). Theory of edge detection. Proceedings of the Royal Society of London. Series B. Biological Sciences, 207(1167), 187–217. https://doi.org/10.1098/rspb.1980.0020
[23] Meester, M. J., & Baslamisli, A. S. (2022). SAR image edge detection: Review and benchmark experiments. International Journal of Remote Sensing, 43(14), 5372–5438. https://doi.org/10.1080/01431161.2022.2131480
[24] Muntarina, K., Mostafiz, R., Khanom, F., Shorif, S. B., & Uddin, M. S. (2023). MultiResEdge: A deep learning-based edge detection approach. Intelligent Systems with Applications, 20, 200274. https://doi.org/10.1016/j.iswa.2023.200274
[25] Navarro, K. (2020). Working in Smoke: Wildfire Impacts on the Health of Firefighters and Outdoor Workers and Mitigation Strategies. Clinics in Chest Medicine, Advances in Occupational and Environmental Lung Diseases, 41(4), 763–769. https://doi.org/10.1016/j.ccm.2020.08.017
[26] Omidi, M., Fatehinya, A., Farahani, M., Akbari, Z., Shahmoradi, S., Yazdian, F., Tahriri, M., Moharamzadeh, K., Tayebi, L., & Vashaee, D. (2017). Characterization of biomaterials. In Biomaterials for Oral and Dental Tissue Engineering (pp. 97–115). Elsevier. https://doi.org/10.1016/B978-0-08-100961-1.00007-4
[27] Otsu, N. (1979). A Threshold Selection Method from Gray-Level Histograms. IEEE Transactions on Systems, Man, and Cybernetics, 9(1), 62–66. https://doi.org/10.1109/TSMC.1979.4310076
[28] Rettenberger, L., Szymanski, N. J., Giunto, A., Dartsi, O., Jain, A., Ceder, G., Hagenmeyer, V., & Reischl, M. (2025). Leveraging unlabeled SEM datasets with self-supervised learning for enhanced particle segmentation. Npj Computational Materials, 11(1), 289. https://doi.org/10.1038/s41524-025-01802-3
[29] Rühle, B., Krumrey, J. F., & Hodoroaba, V.-D. (2021). Workflow towards automated segmentation of agglomerated, non-spherical particles from electron microscopy images using artificial neural networks. Scientific Reports, 11(1), 4942. https://doi.org/10.1038/s41598-021-84287-6
[30] SketchAndCalc. (2024). Area Calculator App | Irregular Area Calculator | Calculate Irregular Shape Area. SketchAndCalc. https://www.sketchandcalc.com/
[31] Sun, W., Duan, F., Zhu, J., Yang, M., & Wang, Y. (2024). An Edge Detection Algorithm for SEM Images of Multilayer Thin Films. Coatings, 14(3). https://doi.org/10.3390/coatings14030313
[32] Wah, W., Gelaw, A., Glass, D. C., Sim, M. R., Hoy, R. F., Berecki-Gisolf, J., & Walker-Bone, K. (2025). Systematic review of impacts of occupational exposure to wildfire smoke on respiratory function, symptoms, measures and diseases. International Journal of Hygiene and Environmental Health, 263, 114463. https://doi.org/10.1016/j.ijheh.2024.114463
[33] Wang, J., Niu, L., Zhang, T. (Tim), & Wang, S. (2019). Detachment of adhered particles from a cloth surface subjected to a rod strike. Aerosol Science and Technology, 53(4), 435–448. https://doi.org/10.1080/02786826.2019.1569199
[34] Wu, Z., Bodla, N., Singh, B., Najibi, M., Chellappa, R., & Davis, L. S. (2019). Soft Sampling for Robust Object Detection. Proceedings of the British Machine Vision Conference (BMVC). https://doi.org/10.48550/arXiv.1806.06986
[35] Xie, S., & Tu, Z. (2015). Holistically-Nested Edge Detection. Proceedings of the IEEE International Conference on Computer Vision (ICCV), 1395–1403. https://doi.org/10.1109/ICCV.2015.164
[36] Zheng, G., Sedlacek, A. J., Aiken, A. C., Feng, Y., Watson, T. B., Raveh-Rubin, S., Uin, J., Lewis, E. R., & Wang, J. (2020). Long-range transported North American wildfire aerosols observed in marine boundary layer of eastern North Atlantic. Environment International, 139, 105680. https://doi.org/10.1016/j.envint.2020.105680
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Sayak Nandi, Md. Momtaz Islam, Roger T. Mailler, Dr. Haejun Park, Adriana Petrova, Lynn M. Boorady, Sumit Mandal (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation.
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
