• Acta Photonica Sinica
  • Vol. 51, Issue 3, 0310005 (2022)
Yunzuo ZHANG1、2、*, Kaina GUO1, Zhaoquan CAI3, and Wenxuan LI1
Author Affiliations
  • 1School of Information Science and Technology,Shijiazhuang Tiedao University,Shijiazhuang 050043,China
  • 2Hebei Key Laboratory of Electromagnetic Environmental Effects and Information Processing,Shijiazhuang Tiedao University,Shijiazhuang 050043,China
  • 3Shanwei Institute of Technology,Shanwei ,Guangdong 516600,China
  • show less
    DOI: 10.3788/gzxb20225103.0310005 Cite this Article
    Yunzuo ZHANG, Kaina GUO, Zhaoquan CAI, Wenxuan LI. Nested Elliptical Spatio-temporal Tubes for Fast Motion Segment Extraction in Surveillance Videos[J]. Acta Photonica Sinica, 2022, 51(3): 0310005 Copy Citation Text show less
    Flowchart of proposed algorithm
    Fig. 1. Flowchart of proposed algorithm
    The formation of nested elliptical spatio-temporal tubes
    Fig. 2. The formation of nested elliptical spatio-temporal tubes
    Spatio-temporal plane maps of nested elliptical spatio-temporal tubes
    Fig. 3. Spatio-temporal plane maps of nested elliptical spatio-temporal tubes
    Schematic diagram of sub-sampling lines and spatio-temporal flow
    Fig. 4. Schematic diagram of sub-sampling lines and spatio-temporal flow
    Spatio-temporal flow of video1
    Fig. 5. Spatio-temporal flow of video1
    The part of the video frames at spatio-temporal flow curve changed of video1
    Fig. 6. The part of the video frames at spatio-temporal flow curve changed of video1
    Spatio-temporal flow analysis chart of video19
    Fig. 7. Spatio-temporal flow analysis chart of video19
    Video sequenceFrame rateResolution ratioFrame number
    Video125384×288566
    Video225720×576543
    Video325320×240944
    Video410384×2883 795
    Video525360×288369
    Viede625360×288336
    Video725360×288289
    Video825320×256137
    Video910384×2881 591
    Video1020352×2882 308
    Video1125352×2884 561
    Video1210384×2882 567
    Video1325352×2881 013
    Video1425720×480892
    Video1525160×120741
    Video1625384×288293
    Video1725384×288381
    Video1825384×2882 358
    Video1930720×4801 332
    Video2030720×480744
    Table 1. Information of experimental videos
    AlgorithmAverage calculating time/sAverage precision /%Average recall/%Average F1 /%
    Pixel difference4614.0992.6482.1187.06
    PMHI61 167.8285.3987.4686.41
    Improved optical flow91 184.8891.7094.3793.02
    Proposed method(1E)33.8299.7385.8292.10
    Proposed method (2E)81.4197.2691.4894.28
    Proposed method (3E)145.4296.4593.8295.12
    Proposed method (4E)199.5396.0194.7595.38
    Table 2. Table for comparing experimental results
    Yunzuo ZHANG, Kaina GUO, Zhaoquan CAI, Wenxuan LI. Nested Elliptical Spatio-temporal Tubes for Fast Motion Segment Extraction in Surveillance Videos[J]. Acta Photonica Sinica, 2022, 51(3): 0310005
    Download Citation