Academia.eduAcademia.edu

Figure 1 – uploaded by Joshan Athanesious J

See full PDFdownloadDownload figure
J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities other points of a trajectory. For example, a vehicle standing in a signal generates repeated data points for a particular time interval (standing time).

Figure 1 J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities other points of a trajectory. For example, a vehicle standing in a signal generates repeated data points for a particular time interval (standing time).

Related Figures (11)

Categorization of Abnormal event detection based on machine learning approach Table 1 J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities
Categorization of Abnormal event detection based on machine learning approach Table 1 J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities
Fig. 3. Estimating the optimal value of eps using Slope calculation. q with respect to Eps and minpts if it satisfies two conditions:
Fig. 3. Estimating the optimal value of eps using Slope calculation. q with respect to Eps and minpts if it satisfies two conditions:
neighborhood points of p to the same cluster if the count is greater than the minpts. Algorithm 2: Clustering using ADBSCAN.
neighborhood points of p to the same cluster if the count is greater than the minpts. Algorithm 2: Clustering using ADBSCAN.
Fig. 5. Object Detection and tracking. (a). Sample Frame. (b) Fore= ground estimation. (c) Vehicle tracking using ID no. (d). Centroid. calculation for each vehicle. J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities
Fig. 5. Object Detection and tracking. (a). Sample Frame. (b) Fore= ground estimation. (c) Vehicle tracking using ID no. (d). Centroid. calculation for each vehicle. J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities
Spatial features along with vehicle ID for i'* frame for time t
Spatial features along with vehicle ID for i'* frame for time t
Fig. 8. Performance of Proposed Method (ADBSCAN) vs NAD —DBSCAN. J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities applied for finding the polygon around the contours while the centroid was calculated using moments. Vehicle Id is given for each moving objects and vehi- cle tracking was done in the next frame using the particular ID. Figure 5 shows the background mod- elling, foreground estimation and vehicle tracking for an input frame. Table 2 shows the feature value such as X, y coordinates and area of the objects at time t.
Fig. 8. Performance of Proposed Method (ADBSCAN) vs NAD —DBSCAN. J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities applied for finding the polygon around the contours while the centroid was calculated using moments. Vehicle Id is given for each moving objects and vehi- cle tracking was done in the next frame using the particular ID. Figure 5 shows the background mod- elling, foreground estimation and vehicle tracking for an input frame. Table 2 shows the feature value such as X, y coordinates and area of the objects at time t.
Comparison results in terms of parameter Eps and Minpts Table 4 Table 3 Frame level comparison of proposed method with state-of-the-art techniques for Chennai Traffic Datasets J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities
Comparison results in terms of parameter Eps and Minpts Table 4 Table 3 Frame level comparison of proposed method with state-of-the-art techniques for Chennai Traffic Datasets J.J. Athanesious et al. / Adaptive density based data mining technique for detection of abnormalities
Performance analysis of ADBSCAN for AVSS dataset positive which is defined as incorrectly identified as anomalies. C f : Count of false negative which gives the missed anomalies. Fig. 10. (a & b) Abnormal event detection in Real life datasets.
Performance analysis of ADBSCAN for AVSS dataset positive which is defined as incorrectly identified as anomalies. C f : Count of false negative which gives the missed anomalies. Fig. 10. (a & b) Abnormal event detection in Real life datasets.
Related topics:
Cognitive ScienceComputer ScienceArtificial Intelligence
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved