(50-4) 08 * << * >> * Русский * English * Содержание * Все выпуски

Алгоритмы компьютерного зрения для обнаружения дыма на видеопоследовательностях
Е.Р. Адамовский1, Р.П. Богуш1, С.В. Абламейко2

1Полоцкий государственный университет имени Евфросинии Полоцкой, 211440, Республика Беларусь, г. Новополоцк, ул. Блохина, д. 29;
2Белорусский государственный университет, 220030, Республика Беларусь, г. Минск, пр. Независимости, д. 4

  Полный текст (PDF)

DOI: 10.18287/COJ1827

ID статьи: 1827

Аннотация:
Представлен аналитический обзор методов и алгоритмов детектирования дыма на последовательностях изображений для раннего обнаружения пожаров с применением систем видеонаблюдения. Рассмотрены визуальные признаки дыма, включая его статические и динамические особенности, которые позволяют отличать его от схожих явлений. Также представлены ограничивающие и мешающие факторы, влияющие на отображение визуальных признаков дыма в реальном мире на видеокадрах. Рассмотрены метрики для оценки эффективности детектирования и локализации областей дыма. Анализируются подходы, основанные на классических алгоритмах компьютерного зрения и на машинном обучении, включая сверточные нейронные сети и трансформеры. Алгоритмы, основанные на машинном обучении, систематизированы с учетом вычислительных затрат. Представлены существующие наборы данных машинного обучения моделей обнаружения дыма, приведены варианты аугментации изображений для расширения обучающей выборки.

Ключевые слова:
детектирование дыма, пространственно-временные признаки дыма, системы видеонаблюдения, компьютерное зрение, нейронные сети.

Цитирование:
Адамовский, Е.Р. Алгоритмы компьютерного зрения для обнаружения дыма на видеопоследовательностях / Е.Р. Адамовский, Р.П. Богуш, С.В. Абламейко // Компьютерная оптика. - 2026. - Т. 50, № 4. - 1827. - doi: 10.18287/COJ1827.

Citation:
Adamovskiy YR, Bohush RP, Ablameyko SV. Computer vision algorithms for detecting smoke in video sequences. Computer Optics 2026; 50(4): 1827. doi: 10.18287/COJ1827.

References:

  1. Fouziya Sulthana S, Wise A, Ravikumar C, Anbazhagan R, Idayachandran G, Pau G. Review study on recent developments in fire sensing methods. IEEE Access 2023; 11: 90269-90282. doi:10.1109/ACCESS.2023.3306812.
  2. Sokolov S, Zhilenkov A, Chernyi S, Nyrkov A. Assessment of the impact of destabilizing factors in the main engine shaft of the adaptive speed controller. Procedia Comput Sci 2018; 125: 420-426. doi:10.1016/j.procs.2017.12.055.
  3. Yang L, Cheng Y, Xu F, Li B, Li X. Real-time smoke detection in surveillance videos using an enhanced RT-DETR framework with triplet attention and HS-FPN. Fire 2024; 7(11): 387. doi:10.3390/fire7110387.
  4. Favorskaya M, Pyataeva A, Popov A. Spatio-temporal smoke clustering in outdoor scenes based on boosted random forests. Procedia Comput Sci 2016; 96: 762-771. doi:10.1016/j.procs.2016.08.231.
  5. Pincott J, Tien PW, Wei S, Calautit JK. Indoor fire detection utilizing computer vision-based strategies. J Build Eng 2022; 61: 105154. doi:10.1016/j.jobe.2022.105154.
  6. Gagliardi A, Saponara S. AdViSED: advanced video smoke detection for real-time measurements in antifire indoor and outdoor systems. Energies 2020; 13(8): 2098. doi:10.3390/en13082098.
  7. Pyataeva AV. Research of methods and development of algorithms for smoke detection in open spaces using video sequences [PhD thesis, in Russian]. Krasnoyarsk: Siberian Federal University; 2016.
  8. Yang Y, Fu T, Song F, Song X, Wang XL, Wang YZ. Wood-burning processes in variable oxygen atmospheres: thermolysis, fire, and smoke release behavior. Polym Degrad Stab 2022; 205: 110158. doi:10.1016/j.polymdegradstab.2022.110158.
  9. Bohush RP, Ablameyko SV. Forest fire smoke detection algorithm in video image [in Russian]. J Belarusian State Univ Math Inform 2021; (1): 91-101. doi:10.33581/2520-6508-2021-1-91-101.
  10. Cao Y, Tang Q, Lu X, Li F, Cao J. STCNet: spatio-temporal cross network for industrial smoke detection. arXiv 2020; arXiv:2011.04863. Available at: https://arxiv.org/abs/2011.04863.
  11. GOST R 70861-2023. Forest protection from fire. Terms and definitions [in Russian]. Moscow: IPK Izdatelstvo Standartov; 2023.
  12. Ye S, Adamovsky Y, Chen H, Bohush R, Ablameyko S. Real-time smoke detection in video based on two-step selection of regions of interest and directional movement analysis. Pattern Recognit Image Anal 2024; 34(4): 1323-1332. doi:10.1134/S1054661824701396.
  13. Carletti V, Greco A, Saggese A, Vento B. A smart visual sensor for smoke detection based on deep neural networks. Sensors 2024; 24(14): 4519. doi:10.3390/s24144519.
  14. Yuan C, Liu Z, Zhang Y. Learning-based smoke detection for unmanned aerial vehicles applied to forest fire surveillance. J Intell Robot Syst 2018; 93(1-2): 337-349. doi:10.1007/s10846-018-0803-y.
  15. Adamovskiy Y, Bohush R. Real-time algorithm for light gray smoke detection in video sequences. Lect Notes Electr Eng 2024; pp. 535-542. doi:10.1007/978-981-97-6934-6_64.
  16. Seefried E. Smoke+: a video dataset for automated fine-grained assessment of smoke opacity [PhD thesis]. Fort Collins, CO: Colorado State University; 2024.
  17. Wang C, Grau A, Guerra E, Shen Z, Hu J, Fan H. Semi-supervised wildfire smoke detection based on smoke-aware consistency. Front Plant Sci 2022; 13: 980425. doi:10.3389/fpls.2022.980425.
  18. Zhao C, Zhao L, Zhang K, Ren Y, Chen H, Sheng Y. Smoke and fire - You only look once: a lightweight deep learning model for video smoke and flame detection in natural scenes. Fire 2025; 8(3): 104. doi:10.3390/fire8030104.
  19. Xu G, Zhang Y, Zhang Q, Lin G, Wang Z, Jia Y, et al. Video smoke detection based on deep saliency network. Fire Saf J 2019; 105: 277-285. doi:10.1016/j.firesaf.2019.03.012.
  20. Favorskaya M, Pyataeva A, Popov A. Verification of smoke detection in video sequences based on spatio-temporal local binary patterns. Procedia Comput Sci 2015; 60: 671-680. doi:10.1016/j.procs.2015.08.205.
  21. Wang Y. Smoke recognition based on machine vision. In: Proc Int Symp Comput Consum Control (IS3C); 2016. pp. 668-671. doi:10.1109/IS3C.2016.172.
  22. Del-Pozo-Velázquez J, Aguiar-Pérez JM, Chamorro-Posada P, Pérez-Juárez MA, Wang X, Casaseca-de-la-Higuera P. Smoke detection in images through fractal dimension-based binary classification. Digit Signal Process 2025; 166: 105346. doi:10.1016/j.dsp.2024.105346.
  23. Toreyin BU, Dedeoglu Y, Cetin AE. Wavelet-based real-time smoke detection in video. In: Proc 13th Eur Signal Process Conf (EUSIPCO); 2005; Antalya, Turkey. pp. 1-4.
  24. Gubenko DD. Study of image filtering algorithms in the problem of improving the quality of smoke detection in a video stream [in Russian]. Izv Tula State Univ Tech Sci 2022; (10): 23-29. doi:10.24412/2071-6168-2022-10-23-29.
  25. Wiegand T, Sullivan GJ, Bjontegaard G, Luthra A. Overview of the H.264/AVC video coding standard. IEEE Trans Circuits Syst Video Technol 2003; 13(7): 560-576.
  26. Hsu YC, Huang TH, Hu TY, Dille P, Prendi S, Hoffman R, et al. Project RISE: recognizing industrial smoke emissions. Proc AAAI Conf Artif Intell 2021; 35(17): 14813-14821.
  27. Gagliardi A, de Gioia F, Saponara S. A real-time video smoke detection algorithm based on Kalman filter and CNN. J Real-Time Image Process 2021; 18(6): 1963-1975. doi:10.1007/s11554-021-01094-y.
  28. Saponara S, Elhanashi A, Gagliardi A. Real-time video fire/smoke detection based on CNN in antifire surveillance systems. J Real-Time Image Process 2020; 18(3): 631-642. doi:10.1007/s11554-020-01044-0.
  29. Dewangan A, Pande Y, Braun HW, Vernon FL, Perez I, Altintas I, et al. FIgLib & SmokeyNet: dataset and deep learning model for real-time wildland fire smoke detection. Remote Sens 2022; 14(4): 1007. doi:10.3390/rs14041007.
  30. Fernandes AM, Utkin AB, Chaves P. Automatic early detection of wildfire smoke with visible light cameras using deep learning and visual explanation. IEEE Access 2022; 10: 12814-12828. doi:10.1109/ACCESS.2022.3145911.
  31. Yazdi A, Qin H, Jordan CB, Yang L, Yan F. Nemo: an open-source transformer-supercharged benchmark for fine-grained wildfire smoke detection. Remote Sens 2022; 14(16): 3979. doi:10.3390/rs14163979.
  32. Chicco D, Jurman G. The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics 2020; 21(1): 6. doi:10.1186/s12864-019-6413-7.
  33. FireWarn: fire hazards detection using deep learning models. IEEE Xplore 2021. Available at: https://ieeexplore.ieee.org/document/9750361. doi:10.1109/CogMI52975.2021.00010.
  34. Huang J, He Z, Guan Y, Zhang H. Real-time forest fire detection by ensemble lightweight YOLOX-L and defogging method. Sensors 2023; 23(4): 1894. doi:10.3390/s23041894.
  35. Huang J, Zhou J, Yang H, Liu Y, Liu H. Small-target forest fire smoke detection model based on deformable transformer for end-to-end object detection. Forests 2023; 14(1): 162. doi:10.3390/f14010162.
  36. Saydirasulovich SN, Mukhiddinov M, Djuraev O, Abdusalomov A, Cho YI. Improved wildfire smoke detection based on YOLOv8 and UAV images. Sensors 2023; 23(20): 8374. doi:10.3390/s23208374.
  37. Hu Y, Zhan J, Zhou G, Chen A, Cai W, Guo K, et al. Fast forest fire smoke detection using MVMNet. Knowl-Based Syst 2022; 241: 108219. doi:10.1016/j.knosys.2022.108219.
  38. Gupta T, Liu H, Bhanu B. Early wildfire smoke detection in videos. In: Proc Int Conf Pattern Recognit (ICPR); 2021. pp. 8523-8530. doi:10.1109/ICPR48806.2021.9413231.
  39. Sheng D, Deng J, Xiang J. Automatic smoke detection based on SLIC-DBSCAN enhanced convolutional neural network. IEEE Access 2021; 9: 63933-63942. doi:10.1109/ACCESS.2021.3075731.
  40. Xia X, Peng Y, Li Z, Shi J, Fang Y. Texture-aware network for enhancing inner smoke representation in visual smoke density estimation. IET Comput Vis 2025; 19(1): 56-70. doi:10.1049/cvi2.70023.
  41. Lin T-Y, Maire M, Belongie S, Hays J, Perona P, Ramanan D, Dollár P, Zitnick CL. Microsoft COCO: Common Objects in Context. In: European Conference on Computer Vision (ECCV); 2014. pp. 740-755. doi:10.1007/978-3-319-10602-1_48.
  42. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: visual explanations from deep networks via gradient-based localization. In: Proc IEEE Int Conf Comput Vis (ICCV); 2017. pp. 618-626. doi:10.1109/ICCV.2017.74.
  43. Li T, Zhao E, Zhang J, Hu C. Detection of wildfire smoke images based on a densely dilated convolutional network. Electronics 2019; 8(10): 1131. doi:10.3390/electronics8101131.
  44. Gragnaniello D, Greco A, Sansone C, Vento B. Fire and smoke detection from videos: a literature review under a novel taxonomy. Expert Syst Appl 2024; 255: 124783. doi:10.1016/j.eswa.2024.124783.
  45. Xu M, Liu T, Zhu K, He Y. Smoke source localization using optical flow imagery in video-based detection. In: Proc 14th Int Congr Image Signal Process Biomed Eng Informatics (CISP-BMEI); 2024. pp. 1-6.
  46. Shi J, Wang W, Gao Y, Yu N. Optimal placement and intelligent smoke detection algorithm for wildfire-monitoring cameras. IEEE Access 2020; 8: 72326-72339. doi:10.1109/ACCESS.2020.2987991.
  47. Fan X, Lei F, Yang K. Real-time detection of smoke and fire in the wild using unmanned aerial vehicle remote sensing imagery. Forests 2025; 16(2): 201. doi:10.3390/f16020201.
  48. Song H, Chen Y. Video smoke detection method based on cell root-branch structure. Signal Image Video Process 2024; 18(5): 4851-4859. doi:10.1007/s11760-024-03121-4.
  49. Ye S, Bai Z, Chen H, Bohush R, Ablameyko S. Effective algorithm to detect both smoke and flame using color and wavelet analysis. Pattern Recognit Image Anal 2017; 27(1): 131-138. doi:10.1134/S1054661817010138.
  50. Bohush R, Adamovskiy Y, Chen H. Real-time smoke detection in video [in Russian]. Herald Polotsk State Univ Ser Fundam Sci 2023; 41(2): 2-8. doi:10.52928/2070-1624-2023-41-2-2-8.
  51. Yuan F, Shi J, Xia X, Fang Y, Fang Z, Mei T. High-order local ternary patterns with locality preserving projection for smoke detection and image classification. Inf Sci 2016; 372: 225-240. doi:10.1016/j.ins.2016.08.040.
  52. Sun Y. Describing motion in computer images' stream: optical flow. In: Computer Vision for Multimedia Applications; 2024. pp. 3-21. doi:10.1007/978-981-99-9191-4_2.
  53. Achanta R, Shaji A, Smith K, Lucchi A, Fua P, Süsstrunk S. SLIC superpixels compared to state-of-the-art superpixel methods. IEEE Trans Pattern Anal Mach Intell 2012; 34(11): 2274-2282. doi:10.1109/TPAMI.2012.120.
  54. Rehman SU, Asghar S, Fong S, Sarasvady S. DBSCAN: past, present and future. In: Proc 5th Int Conf Appl Digit Inf Web Technol (ICADIWT); 2014. doi:10.1109/ICADIWT.2014.6814687.
  55. Vaishnav P, Sankarasubramanian P. Realtime pipeline fire and smoke detection using a lightweight CNN model. In: Proc Int Conf Mach Learn Appl Neural Technol (ICMLANT); 2021. doi:10.1109/ICMLANT53170.2021.9690545.
  56. Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. arXiv 2017; arXiv:1706.03762. Available at: https://arxiv.org/abs/1706.03762.
  57. Jiang P, Ergu D, Liu F, Cai Y, Ma B. Review of YOLO algorithm developments. Procedia Comput Sci 2022; 199: 1066-1073. doi:10.1016/j.procs.2022.01.135.
  58. Özel B, Alam MS, Khan MU. Review of modern forest fire detection techniques: innovations in image processing and deep learning. Information 2024; 15(9): 538. doi:10.3390/info15090538.
  59. Shi X, Lu N, Cui Z. Smoke detection based on dark channel and convolutional neural networks. In: Proc Big Data Intell Anal Conf (BigDIA); 2019. pp. 23-28. doi:10.1109/BigDIA.2019.8802668.
  60. Jadon A, Omama M, Varshney A, Ansari MS, Sharma R. FireNet: a specialized lightweight fire and smoke detection model for real-time IoT applications. arXiv 2019; arXiv:1905.11922. Available at: https://arxiv.org/abs/1905.11922.
  61. FireNET: deep learning model for detecting fire in video and camera streams. Available at: https://github.com/OlafenwaMoses/FireNET.
  62. Khodarahmi M, Maihami V. Review on Kalman filter models. Arch Comput Methods Eng 2022; 30: 727-747. doi:10.1007/s11831-022-09815-7.
  63. FireSense dataset. Available at: https://www.kaggle.com/datasets/chrisfilo/firesense.
  64. Filonenko A, Hernandez DC, Jo KH. Fast smoke detection for video surveillance using CUDA. IEEE Trans Ind Inform 2018; 14(2): 725-733. doi:10.1109/TII.2017.2757457.
  65. Li H, Li J, Wei H, Liu Z, Zhan Z, Ren Q. Slim-neck by GSConv: a better design paradigm of detector architectures for autonomous vehicles. J Real-Time Image Process 2024; in press. doi:10.1007/s11554-024-01436-6. Available at: https://arxiv.org/abs/2206.02424.
  66. Zhu L, Wang X, Ke Z, Zhang W, Lau R. BiFormer: vision transformer with bi-level routing attention. In: Proc IEEE Conf Comput Vis Pattern Recognit (CVPR); 2023. doi:10.1109/CVPR52729.2023.00995.
  67. Han Y, Li B, Hu Y, et al. LUFFD-YOLO: A lightweight model for UAV remote sensing forest fire detection based on attention mechanism and multi-level feature fusion. Remote Sens 2024; 16(12): 2177. doi:10.3390/rs16122177.
  68. Wang G. Multiple scenarios, multiple weather conditions, multiple lighting conditions and multiple wildfire objects synthetic forest wildfire dataset (M4SFWD). IEEE Dataport 2024. doi:10.21227/m9kz-bw61. Available at: https://ieee-dataport.org/documents/multiple-scenarios-multiple-weather-conditions-multiple-lighting-conditions-and-multiple.
  69. Fire-Smoke-Dataset. Available at: https://github.com/DeepQuestAI/Fire-Smoke-Dataset.
  70. Carion N, Massa F, Synnaeve G, Usunier N, Kirillov A, Zagoruyko S. End-to-end object detection with transformers. arXiv 2020; arXiv:2005.12872. Available at: https://arxiv.org/abs/2005.12872.
  71. HPWREN FIgLib data. Available at: https://cdn.hpwren.ucsd.edu/HPWREN-FIgLib-Data/index.html.
  72. ALERT Wildfire camera network. Available at: https://www.alertwildfire.org/.
  73. Zhu X, Su W, Lu L, Li B, Wang X, Dai J. Deformable DETR: deformable transformers for end-to-end object detection. arXiv 2021; arXiv:2010.04159. Available at: https://arxiv.org/abs/2010.04159.
  74. Yuan L, Chen Y, Wang T, Yu W, Shi Y, Francis J, et al. Tokens-to-token ViT: training vision transformers from scratch on ImageNet. arXiv 2021; arXiv:2101.11986. Available at: https://arxiv.org/abs/2101.11986.
  75. Huo Y, Zhang Q, Jia Y, Liu D, Guan J, Lin G, et al. Deep separable convolutional neural network for multiscale image-based smoke detection. Fire Technol 2022; 58(3): 1445-1468. doi:10.1007/s10694-021-01199-7.
  76. Yang C, Pan Y, Cao Y, Lu X. CNN-Transformer hybrid architecture for early fire detection. Lect Notes Comput Sci 2022; pp. 570-581. doi:10.1007/978-3-031-15937-4_48.
  77. Tao H, Duan Q. Adaptive frame selection network with enhanced dilated convolution for video smoke recognition. Expert Syst Appl 2023; 225: 119371. doi:10.1016/j.eswa.2022.119371.
  78. Chen X, An Q, Yu K, Ban Y. Novel fire identification algorithm based on improved color segmentation and enhanced feature data. IEEE Trans Instrum Meas 2021; 70: 1-15. doi:10.1109/TIM.2021.3075380.
  79. Vinícius P, Campos RJ, Rezende TM, Lisboa AC, Barbosa AV. Hybrid method for fire detection based on spatial and temporal patterns. Neural Comput Appl 2023; 35(13): 9349-9361. doi:10.1007/s00521-023-08260-2.
  80. Huo Y, Zhang Q, Zhang Y, Zhu J, Wang J. 3DVSD: an end-to-end 3D convolutional object detection network for video smoke detection. Fire Saf J 2022; 134: 103690. doi:10.1016/j.firesaf.2022.103690.
  81. Cao Y, Yang F, Tang Q, Lu X. Attention-enhanced bidirectional LSTM for early forest fire smoke recognition. IEEE Access 2019; 7: 154732-154742. doi:10.1109/ACCESS.2019.2946712.
  82. Heusel M, Ramsauer H, Unterthiner T, Nessler B, Hochreiter S. GANs trained by a two time-scale update rule converge to a local Nash equilibrium. arXiv 2017; arXiv:1706.08500. Available at: https://arxiv.org/abs/1706.08500.
  83. Xu G, Zhang Q, Liu D, Li G, Wang JJ, Zhang Y. Adversarial adaptation from synthesis to reality in fast detector for smoke detection. IEEE Access 2019; 7: 29471-29483. doi:10.1109/ACCESS.2019.2902606.
  84. Tong Z, Chen Y, Xu Z, Yu R. Wise-IoU: bounding box regression loss with dynamic focusing mechanism. arXiv 2023; arXiv:2301.10051. Available at: https://arxiv.org/abs/2301.10051.
  85. Gevorgyan Z. SIoU loss: more powerful learning for bounding box regression. arXiv 2022; arXiv:2205.12740. Available at: https://arxiv.org/abs/2205.12740.
  86. dsw2017 smoke dataset. Available at: https://github.com/mtcazzolato/dsw2017.
  87. VisiFire demo videos. Available at: http://signal.ee.bilkent.edu.tr/VisiFire/Demo.
  88. D-Fire videos. Available at: https://clk.li/BhNw.
  89. MIVIA smoke dataset. Available at: https://mivia.unisa.it/datasets/video-analysis-datasets/smoke-detection-dataset/.
  90. furg-fire-dataset. Available at: https://github.com/steffensbola/furg-fire-dataset.
  91. Computer Vision and Pattern Recognition Laboratory homepage. Available at: https://cvpr.kmu.ac.kr/.
  92. ImageNet dataset. Available at: https://image-net.org/.

Россия, 443001, Самара, ул. Молодогвардейская, 151; электронная почта: journal@computeroptics.ru; тел: +7 (846) 242-41-24 (ответственный секретарь), +7 (846) 332-56-22 (технический редактор), факс: +7 (846) 332-56-20