• Laser & Optoelectronics Progress
  • Vol. 60, Issue 12, 1210021 (2023)
Hechao Yang, Gang Chen, and Chunyu Yu*
Author Affiliations
  • College of Electronic and Optical Engineering & College of Microelectronics, Nanjing University of Posts and Telecommunications, Nanjing 210023, Jiangsu, China
  • show less
    DOI: 10.3788/LOP221654 Cite this Article Set citation alerts
    Hechao Yang, Gang Chen, Chunyu Yu. Moving Target Detection Algorithm Based on New Background Extraction[J]. Laser & Optoelectronics Progress, 2023, 60(12): 1210021 Copy Citation Text show less
    Analysis of pixel change in background area. (a) Background analysis diagram; (b) change rule of pixel value at a certain position in three regions
    Fig. 1. Analysis of pixel change in background area. (a) Background analysis diagram; (b) change rule of pixel value at a certain position in three regions
    Inter frame pixel difference rule of 150 consecutive frames
    Fig. 2. Inter frame pixel difference rule of 150 consecutive frames
    Effect of H on image quality. (a) highway; (b) skating; (c) snowfall; (d) pedestrians
    Fig. 3. Effect of H on image quality. (a) highway; (b) skating; (c) snowfall; (d) pedestrians
    Flow chart of proposed algorithm
    Fig. 4. Flow chart of proposed algorithm
    Background reconstruction images. (a) highway; (b) skating; (c) snowfall; (d) pedestrians
    Fig. 5. Background reconstruction images. (a) highway; (b) skating; (c) snowfall; (d) pedestrians
    Relationship between reconstructed background quality and the number of frames involved in reconstruction
    Fig. 6. Relationship between reconstructed background quality and the number of frames involved in reconstruction
    Moving target detection. (a) highway; (b) skating; (c) snowfall; (d) pedestrians
    Fig. 7. Moving target detection. (a) highway; (b) skating; (c) snowfall; (d) pedestrians
    Evaluation indexAlgorithmhighwayskatingsnowfallpedestrians
    precisionViBe0.90920.90830.10130.8455
    reference[60.90930.91420.07530.8391
    GMM0.41260.69050.29100.5794
    proposed algorithm0.90990.98660.83280.9685
    recallViBe0.82200.75870.41240.8902
    reference[60.82290.76480.34450.8917
    GMM0.41260.22740.09500.4349
    proposed algorithm0.91000.80280.82170.9318
    F1-measureViBe0.86340.82680.16260.8673
    reference[60.86400.83290.12360.8646
    GMM0.56210.34210.14320.4968
    proposed algorithm0.90990.88530.82720.9498
    Table 1. Algorithm performance index
    Evaluation indexAlgorithmhighwayskatingsnowfallpedestrians
    FPRViBe0.00520.00500.07760.0018
    reference[60.00520.00470.08880.0018
    GMM0.00350.00660.00420.0034
    proposed algorithm0.00570.00070.00300.0003
    Table 2. Anti noise index