Lightning, as a powerful natural discharge phenomenon, is harmful to human production and life, natural ecology, railroad transportation, etc. The study of lightning is of great significance, and image segmentation is the key link in lightning research. In this paper, a method of lightning image segmentation based on contour detection and curvature analysis is proposed. This method can accurately segment the lightning channel and locate the coordinates of the lightning strike point, which is of great significance for the in-depth study of lightning. Meanwhile, this project collects lightning images from multiple sources to construct a dataset. It realizes lightning channel extraction and strike point localization through the steps of image preprocessing, thresholding and contour detection, curvature analysis, and lightning strike point recognition. The experimental results show that the method is effective in single-branch and multi-branch lightning and cloud flash image segmentation, with obvious advantages over traditional methods, and the strike point recognition algorithm has high reliability.

1. [1] H. H. Goh et al., “A review of lightning protection system – Risk assessment and application,” Indonesian J. of Electrical Engineering and Computer Science, Vol.8, No.1, pp. 221-229, 2017. https://doi.org/10.11591/ijeecs.v8.i1.pp221-229
2. [2] Y. Zhang and X. Zhou, “Review and progress of lightning research,” J. of Applied Meteorological Science, Vol.17, No.6, pp. 829-834, 2006 (in Chinese).
3. [3] A. Srivastava et al., “Lightning nowcasting with an algorithm of thunderstorm tracking based on lightning location data over the Beijing area,” Advances in Atmospheric Sciences, Vol.39, No.1, pp. 178-188, 2022. https://doi.org/10.1007/s00376-021-0398-2
4. [4] Q. Xu, “Analysis and optimization of equipment failure caused by lightning stroke of the tower equipped with video monitoring system along the railway,” Railway Signalling & Communication, Vol.59, No.4, pp. 53-57, 2023 (in Chinese). https://doi.org/10.13879/j.issn.1000-7458.2023-04.22262
5. [5] G. V. Lopez, J. Orejudos, and A. A. Dellagas, “ULAT: Deployment of dense and nationwide lightning detection network for weather forecasting in the Philippines,” 2022 IEEE Global Humanitarian Technology Conf., pp. 54-61, 2022. https://doi.org/10.1109/ghtc55712.2022.9911055
6. [6] J. Bao et al., “Lightning performance evaluation of transmission line based on data-driven lightning identification, tracking, and analysis,” IEEE Trans. on Electromagnetic Compatibility, Vol.63, No.1, pp. 160-171, 2021. https://doi.org/10.1109/TEMC.2020.2988902
7. [7] H. Chen, R. Li, D. Yang, and T. Xiao, “Research and implementation of lightning monitoring technology in railway stations and yards based on optical imaging,” Railway Signalling & Communication, Vol.59, No.3, pp. 14-17+23, 2023 (in Chinese). https://doi.org/10.13879/j.issn.1000-7458.2023-03.22377
8. [8] X. Yang et al., “Applications of 3 threshold methods to the lightning channel image recognition,” J. of Applied Meteorological Science, Vol.25, No.4, pp. 427-435, 2014 (in Chinese).
9. [9] F. Nie, M. Liu, and P. Zhang, “Multilevel thresholding with divergence measure and improved particle swarm optimization algorithm for crack image segmentation,” Scientific Reports, Vol.14, No.1, Article No.7642, 2024. https://doi.org/10.1038/s41598-024-58456-2
10. [10] M. Nazarkevych, M. Kostiak, N. Oleksiv, V. Vysotska, and A.-T. Shvahuliak, “A YOLO-based method for object contour detection and recognition in video sequences,” Proc. of the Workshop Cybersecurity Providing in Information and Telecommunication Systems (CPITS 2024), pp. 49-58, 2024.
11. [11] Q. Lei, J. Zhong, C. Wang, Q. Dai, and R. Li, “Adaptive thresholding based on multi-task learning for refining binary medical image segmentation,” 2023 IEEE Int. Conf. on Bioinformatics and Biomedicine, pp. 3059-3066, 2023. https://doi.org/10.1109/BIBM58861.2023.10385508
12. [12] B. Yang, W. Liu, S. Lu, and J. Luo, “Feature extraction of lubricating oil debris signal based on segmentation entropy with an adaptive threshold,” Sensors, Vol.24, No.5, Article No.1380, 2024. https://doi.org/10.3390/s24051380
13. [13] Z. Xie, J. Wu, W. Tang, and Y. Liu, “Advancing image segmentation with DBO-Otsu: Addressing rubber tree diseases through enhanced threshold techniques,” PLOS ONE, Vol.19, No.3, Article No.e0297284, 2024. https://doi.org/10.1371/journal.pone.0297284
14. [14] Z. Cai, G. Li, J. Zhang, and S. Xiong, “Using an artificial Physarum polycephalum colony for threshold image segmentation,” Applied Sciences, Vol.13, No.21, Article No.11976, 2023. https://doi.org/10.3390/app132111976
15. [15] C. Zeng, Y. Li, K. Yang, and C. Li, “Contour detection based on a non-classical receptive field model with butterfly-shaped inhibition subregions,” Neurocomputing, Vol.74, No.10, pp. 1527-1534, 2011. https://doi.org/10.1016/j.neucom.2010.12.022
16. [16] B. Liu, “Long-distance recognition of crack width in building wall based on binocular vision,” 2021 3rd Int. Conf. on Artificial Intelligence and Advanced Manufacture, pp. 1908-1912, 2021. https://doi.org/10.1145/3495018.3495513
17. [17] T. He, H. Li, Z. Qian, C. Niu, and R. Huang, “Research on weakly supervised pavement crack segmentation based on defect location by generative adversarial network and target re‐optimization,” Construction and Building Materials, Vol.411, Article No.134668, 2024. https://doi.org/10.1016/j.conbuildmat.2023.134668
18. [18] L. Zhao and R. Zhu, “Research on image contour edge analysis based on canny edge detector,” Academic J. of Computing & Information Science, Vol.5, No.1, pp. 70-75, 2022. https://doi.org/10.25236/AJCIS.2022.050114
19. [19] H. Henry and M. Adda-Bedia, “Fractographic aspects of crack branching instability using a phase-field model,” Physical Review E, Vol.88, No.6, Article No.060401, 2013. https://doi.org/10.1103/PhysRevE.88.060401
20. [20] M.-R. Siadat, K. Elisevich, H. Soltanian-Zadeh, A. Eetemadi, and B. Smith, “Curvature analysis of perisylvian epilepsy,” Acta Neurologica Belgica, Vol.123, No.6, pp. 2303-2313, 2023. https://doi.org/10.1007/s13760-023-02238-6
21. [21] L. M. Harrison et al., “Three-dimensional assessment of frontal bossing and temporal pinching in patients with sagittal craniosynostosis using curvature analysis,” Plastic and Reconstructive Surgery, Vol.152, No.3, pp. 603-610, 2023. https://doi.org/10.1097/PRS.0000000000010277
22. [22] Y. S. Alsahafi, D. S. Elshora, E. R. Mohamed, and K. M. Hosny, “Multilevel threshold segmentation of skin lesions in color images using coronavirus optimization algorithm,” Diagnostics, Vol.13, No.18, Article No.2958, 2023. https://doi.org/10.3390/diagnostics13182958
23. [23] A. Hiuchi and T. Ogawa, “Contour detection methods using random forest for augmented reality,” Abstracts of IEICE Trans. on Information and Systems D (Japanese Edition), Vol.J98-D, No.9, pp. 1201-1211, 2015.
24. [24] Xerox Corporation, “Patent issued for automated contour detection methods, systems and processor-readable media,” J. of Engineering, 2015.
25. [25] G. Wang and B. De Baets, “Contour detection based on anisotropic edge strength and hierarchical superpixel contrast,” Signal, Image and Video Processing, Vol.13, No.8, pp. 1657-1665, 2019. https://doi.org/10.1007/s11760-019-01517-1
26. [26] Z. Zhang, H. Wang, G. Yu, and J. Zhao, “Research on four-point air bending process and contour detection method for JCO forming process of LSAW pipes,” Metals, Vol.9, No.8, Article No.859, 2019. https://doi.org/10.3390/met9080859
27. [27] C. Li, V. T. Chau, H. Xie, and Z. P. Bažant, “Recent advances in mechanics of fracking and new results on 2D simulation of crack branching in anisotropic gas or oil shale,” Acta Mechanica, Vol.229, No.2, pp. 975-992, 2018. https://doi.org/10.1007/s00707-017-2010-5
28. [28] F. Sokhangou, L. Sorelli, L. Chouinard, P. Dey, and D. Conciatori, “Detecting multiple damages in UHPFRC beams through modal curvature analysis,” Sensors, Vol.24, No.3, Article No.971, 2024. https://doi.org/10.3390/s24030971
29. [29] G. Chen et al., “Geological structure identification of coalbed methane reservoir based on trend surface and curvature analysis algorithms,” Earth Science Informatics, Vol.17, No.2, pp. 1345-1358, 2024. https://doi.org/10.1007/s12145-024-01232-7
30. [30] A. A. Kalitzeos, G. Y. H. Lip, and R. Heitmar, “Retinal vessel tortuosity measures and their applications,” Experimental Eye Research, Vol.106, pp. 40-46, 2013. https://doi.org/10.1016/j.exer.2012.10.015
31. [31] D. Orozco-Gomez, F. Bolanos, J. Herrera-Murcia, and A. Espinosa-Bedoya, “A classification model for lightning images without branches based on the tortuosity metric,” Modeling Earth Systems and Environment, Vol.10, No.2, pp. 2447-2461, 2024. https://doi.org/10.1007/s40808-023-01900-5
32. [32] Z. Qian et al., “Lightning identification method based on deep learning,” Atmosphere, Vol.13, No.12, Article No.2112, 2022. https://doi.org/10.3390/atmos13122112