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

Acta Optica Sinica, Volume. 38, Issue 8, 0815007(2018)

Anomaly Detection and Location in Crowded Surveillance Videos

Peipei Zhou1,2,3,4、*, Qinghai Ding1,5、*, Haibo Luo1,3,4, and Xinglin Hou1,2,3,4
Author Affiliations
  • 1 Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, Liaoning 110016
  • 2 University of Chinese Academy of Sciences, Beijing 100049
  • 3 Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang, Liaoning 110016
  • 4 Key Laboratory of Image Understanding and Computer Vision, Shenyang, Liaoning 110016
  • 5 Space Star Technology Co., Ltd., Beijing 100086
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (11)
    Equations (0)
    References (33)
    Cited By (6)
    Tools
    Paper Information
    Topics
    Figures & Tables(11)
    Flow chart of abnormal behavior detection and location

    Fig. 1. Flow chart of abnormal behavior detection and location

    Download full size
    Segmented motion-region images. (a) Original image; (b) motion regions achieved by the ViBE method; (c) image with motion magnitude; (d) motion regions achieved by the proposed algorithm

    Fig. 2. Segmented motion-region images. (a) Original image; (b) motion regions achieved by the ViBE method; (c) image with motion magnitude; (d) motion regions achieved by the proposed algorithm

    Download full size
    Filtering results according to motion continuity for the abnormal regions which are predicted by the classifiers

    Fig. 3. Filtering results according to motion continuity for the abnormal regions which are predicted by the classifiers

    Download full size
    Examples of normal and abnormal crowed activities in UMN dataset. (a) Normal behaviors; (b) abnormal behaviors

    Fig. 4. Examples of normal and abnormal crowed activities in UMN dataset. (a) Normal behaviors; (b) abnormal behaviors

    Download full size
    ROC curves of different methods in two criterions on the UCSD ped2 dataset. (a) Frame-level; (b) pixel-level

    Fig. 5. ROC curves of different methods in two criterions on the UCSD ped2 dataset. (a) Frame-level; (b) pixel-level

    Download full size
    Pixel-level detection results with different methods on the UCSD ped2 dataset. (a) Social Force; (b) MPPCA; (c) T-MDT; (d) S-MDT; (e) proposed method

    Fig. 6. Pixel-level detection results with different methods on the UCSD ped2 dataset. (a) Social Force; (b) MPPCA; (c) T-MDT; (d) S-MDT; (e) proposed method

    Download full size
    ROC curves of different methods for frame-level detection on the UMN dataset

    Fig. 7. ROC curves of different methods for frame-level detection on the UMN dataset

    Download full size
    Detection and location results of three different scenes on UMN dataset

    Fig. 8. Detection and location results of three different scenes on UMN dataset

    Download full size
    Relationship curves between the parameter k and the system performance on two datasets

    Fig. 9. Relationship curves between the parameter k and the system performance on two datasets

    Download full size

    • Table 1. Comparison results of different methods on the UCSD ped2 dataset for frame-level and pixel-level detection

      View table

      Table 1. Comparison results of different methods on the UCSD ped2 dataset for frame-level and pixel-level detection

      MethodFrame-level criterionPixel-level criterion
      EER /%AUCDR /%AUCTime /s
      Social Force[33]42.00.70227.60.217-
      MPPCA[34]31.10.71022.40.222-
      S-MDT[32]28.70.75063.40.6650.69
      T-MDT[32]27.90.76556.80.5220.64
      ST-CNN[23]24.40.86081.90.8800.37
      Motion Energy[16]22.00.81055.00.5800.08
      Proposed10.30.90589.70.9020.26

    • Table 2. Comparison results of different methods on the UMN dataset in AUC and EER criterion for frame-level detection

      View table

      Table 2. Comparison results of different methods on the UMN dataset in AUC and EER criterion for frame-level detection

      MethodAUCEER /%
      Social Force[33]0.94912.6
      Sparse[1]0.9962.8
      H-MDT CRF[19]0.9953.7
      ST-CNN[23]0.9963.3
      Motion Energy[16]0.9894.1
      Proposed0.9892.7
    Tools

    Get Citation

    Copy Citation Text

    Peipei Zhou, Qinghai Ding, Haibo Luo, Xinglin Hou. Anomaly Detection and Location in Crowded Surveillance Videos[J]. Acta Optica Sinica, 2018, 38(8): 0815007

    Download Citation

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

    Category: Machine Vision

    Received: Jan. 22, 2018

    Accepted: Feb. 26, 2018

    Published Online: Sep. 6, 2018

    The Author Email:

    DOI:10.3788/AOS201838.0815007

    Topics

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