Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Chinese Journal of Lasers, Volume. 48, Issue 3, 0309001(2021)

Statistical-Based Adaptive Background Modeling Algorithm for Grayscale Video

Jiawen Wu1,2 and Shiyong Wang1、*
Author Affiliations
  • 1Key Laboratory of Infrared System Detection and Imaging Technology, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, China
  • 2University of Chinese Academy of Sciences, Beijing 100049, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (10)
    Equations (0)
    References (24)
    Cited By (6)
    Tools
    Paper Information
    Topics
    Figures & Tables(10)
    Background modeling results of different algorithms. (a) Original images; (b) ViBe algorithm; (c) GMM algorithm; (d) proposed algorithm

    Fig. 1. Background modeling results of different algorithms. (a) Original images; (b) ViBe algorithm; (c) GMM algorithm; (d) proposed algorithm

    Download full size
    Process of noise estimation by three-frame differential transformation algorithm. (a) Input image sequence; (b) grayscale change curve of sequence A; (c) grayscale change curve of sequence B; (d) frequency change curves of sequence A and sequence B

    Fig. 2. Process of noise estimation by three-frame differential transformation algorithm. (a) Input image sequence; (b) grayscale change curve of sequence A; (c) grayscale change curve of sequence B; (d) frequency change curves of sequence A and sequence B

    Download full size
    Experimental results of dynamic or static background under different scenes. (a) Original image of static background; (b) gray histogram of Fig. (a) box 1; (c) gray histogram of Fig. (a) box 2; (d) original image of dynamic background; (e) gray histogram of Fig. (d) box 1; (f) Figure (d) gray histogram of Fig. (d) box 2

    Fig. 3. Experimental results of dynamic or static background under different scenes. (a) Original image of static background; (b) gray histogram of Fig. (a) box 1; (c) gray histogram of Fig. (a) box 2; (d) original image of dynamic background; (e) gray histogram of Fig. (d) box 1; (f) Figure (d) gray histogram of Fig. (d) box 2

    Download full size
    Dynamic and static background partition results under different scenes.(a) PETS2006 image; (b) freeway scene; (c) lake scene; (d) division result of Fig. (a); (e) division result of Fig. (b); (f) division result of Fig. (c)

    Fig. 4. Dynamic and static background partition results under different scenes.(a) PETS2006 image; (b) freeway scene; (c) lake scene; (d) division result of Fig. (a); (e) division result of Fig. (b); (f) division result of Fig. (c)

    Download full size
    Target extraction process under dynamic background. (a) Original image; (b) 1st level threshold; (c) 2nd level threshold; (d) elimination of false detection; (e) morphological operation

    Fig. 5. Target extraction process under dynamic background. (a) Original image; (b) 1st level threshold; (c) 2nd level threshold; (d) elimination of false detection; (e) morphological operation

    Download full size
    Process of light adjustment algorithm and corresponding grayscale straight maps. (a) Background image of 297th frame; (b) original image of 297th frame; (c) set of pixels whose gray value is equal to GB; (d) final results

    Fig. 6. Process of light adjustment algorithm and corresponding grayscale straight maps. (a) Background image of 297th frame; (b) original image of 297th frame; (c) set of pixels whose gray value is equal to GB; (d) final results

    Download full size
    Results of each stage of foreground shadow elimination method. (a) Original image; (b) foreground mask; (c) foreground to background ratio; (d) ratio gradient; (e) gradient graph binarization; (f) convex hull of point set; (g) convex hull intersects foreground

    Fig. 7. Results of each stage of foreground shadow elimination method. (a) Original image; (b) foreground mask; (c) foreground to background ratio; (d) ratio gradient; (e) gradient graph binarization; (f) convex hull of point set; (g) convex hull intersects foreground

    Download full size
    Processing results of different algorithms. (a) Original images; (b) Truth graphs; (c) CodeBook algorithm; (d) GMM algorithm; (e) KDE algorithm; (f) PBAS algorithm; (g) ViBe algorithm; (h) proposed algorithm

    Fig. 8. Processing results of different algorithms. (a) Original images; (b) Truth graphs; (c) CodeBook algorithm; (d) GMM algorithm; (e) KDE algorithm; (f) PBAS algorithm; (g) ViBe algorithm; (h) proposed algorithm

    Download full size
    Processing results of mid-wave infrared image by different algorithms. (a) Original images; (b) Truth graphs; (c) CodeBook algorithm; (d) GMM algorithm; (e) KDE algorithm; (f) PBAS algorithm; (g) ViBe algorithm; (h) proposed algorithm

    Fig. 9. Processing results of mid-wave infrared image by different algorithms. (a) Original images; (b) Truth graphs; (c) CodeBook algorithm; (d) GMM algorithm; (e) KDE algorithm; (f) PBAS algorithm; (g) ViBe algorithm; (h) proposed algorithm

    Download full size

    • Table 1. Evaluation index results of different algorithms

      View table

      Table 1. Evaluation index results of different algorithms

      AlgorithmImage 1Image 2Image 3
      RePrPPWCFRePrPPWCFRePrPPWCF
      CodeBook0.7400.7090.7290.7240.7730.6393.8350.7000.8770.2872.8170.433
      GMM0.7860.6930.7260.7370.7620.7223.0680.7410.9050.5481.0310.683
      KDE0.8700.6950.6610.7730.6990.4666.3630.5590.6590.8040.6150.724
      PBAS0.6510.6670.8720.6590.8120.7142.9670.7600.9630.5001.2230.658
      ViBe0.6000.6620.9140.6300.7470.7123.2040.7290.9250.5391.0620.681
      Proposed algorithm0.9680.6980.5830.8110.9890.6972.5490.8170.9600.5381.0570.690
      AlgorithmImage 4Image 5Image 6
      RePrPPWCFRePrPPWCFRePrPPWCF
      CodeBook0.9320.4129.5860.5710.7280.6523.6370.6880.5190.7011.6390.597
      GMM0.4450.6375.5370.5240.6950.5844.4000.6350.5070.7251.6000.597
      KDE0.4140.5856.0280.4850.2940.6444.7760.4030.2450.7771.9280.372
      PBAS0.3360.5136.7340.4060.3120.8664.0470.4590.6410.6721.5710.656
      ViBe0.1620.6136.4440.2560.4410.5555.0220.4910.6050.6831.5780.642
      Proposed algorithm0.9280.8052.0370.8620.8580.7752.1500.8150.6710.7291.3530.699
    Tools

    Get Citation

    Copy Citation Text

    Jiawen Wu, Shiyong Wang. Statistical-Based Adaptive Background Modeling Algorithm for Grayscale Video[J]. Chinese Journal of Lasers, 2021, 48(3): 0309001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: holography and information processing

    Received: Jun. 29, 2020

    Accepted: Sep. 9, 2020

    Published Online: Feb. 8, 2021

    The Author Email: Wang Shiyong (wangshiyong@mail.sitp.ac.cn)

    DOI:10.3788/CJL202148.0309001

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology