Advanced Vision - TASK 6 MATLAB CODE

This task tracks a falling ball.

These files are used as part of TASK 6:

There are 4 separate demos:

  1. The first just does detection by background subtraction (enter matab command: detect).
    This can be considered as the ground truth.
    Example [MPG]
  2. The second feeds the detection output into a Kalman filter (enter matab command: kalman). The predicted position from the kalman filter (red) is compared against the actual ground truth position (green).
    Example [MPG]
  3. The third wraps a condensation filter about the Kalman filter (enter matab command: condense). The predicted position from the condensation filter (red) is compared against the actual ground truth position (green).
    Example [MPG]
  4. The fourth uses non-parametric background modeling to identify outlying (foreground) pixels (enter matab command: detectp).

Each demo displays the tracked ball sequence (with predicted position against true position for kalman/condense) and a plot of ball height in the image over time measured in frames of the sequence (for background subtraction & kalman demos).

Example mpeg movies can be viewed using the gmplayer command.

All 3 demos expect the Ball Video Frames in ./DATA/*

Further demos of the system 6 ball and change detection examples.

[AV index page]
This page is maintained by the course lecturer, Toby Breckon, toby.breckon@ed.ac.uk, room 1104B JCMB, ext 513448


Home : Teaching : Courses : Av : MATLAB 

Informatics Forum, 10 Crichton Street, Edinburgh, EH8 9AB, Scotland, UK
Tel: +44 131 651 5661, Fax: +44 131 651 1426, E-mail: school-office@inf.ed.ac.uk
Please contact our webadmin with any comments or corrections. Logging and Cookies
Unless explicitly stated otherwise, all material is copyright © The University of Edinburgh