Smoke segmentation is a critical task in computer vision, with applications in disaster response, environmental monitoring, and safety assessment, etc. This study presents an efficient method for segmenting smoke using KNN-based background modeling for foreground extraction and the SegFormer semantic segmentation algorithm. The KNN background modeling effectively isolates the explosion smoke regions, generating a large dataset with minimal manual annotation workload. We compare the performance of several semantic segmentation algorithms, namely UNet, DeepLab v3 plus, and SegFormer, using the generated dataset. Our results demonstrate that SegFormer outperforms the other algorithms in terms of accuracy and efficiency. The proposed method not only simplifies the dataset generation process but also provides accurate segmentation of smoke, making it a valuable tool for real-world applications in disaster response and environmental monitoring.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Effective Smoke Segmentation Using KNN Background Modeling and SegFormer


    Additional title:

    Lect. Notes Electrical Eng.


    Contributors:
    Long, Shengzhao (editor) / Dhillon, Balbir S. (editor) / Yang, Junjie (author) / Liu, Haoting (author) / Lv, Meng (author) / Wang, Mengmeng (author) / Li, Qing (author)

    Conference:

    International Conference on Man-Machine-Environment System Engineering ; 2023 ; Beijing, China October 20, 2023 - October 23, 2023



    Publication date :

    2023-09-05


    Size :

    7 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English




    Semantic Segmentation of Unmanned Aerial Vehicle Remote Sensing Images using SegFormer

    Spasev, Vlatko / Dimitrovski, Ivica / Chorbev, Ivan et al. | ArXiv | 2024

    Free access

    Semantic Segmentation of Unmanned Aerial Vehicle Remote Sensing Images Using SegFormer

    Spasev, Vlatko / Dimitrovski, Ivica / Chorbev, Ivan et al. | Springer Verlag | 2025



    An Adaptive Background Modeling Method for Foreground Segmentation

    Zhong, Zuofeng / Zhang, Bob / Lu, Guangming et al. | IEEE | 2017