detectforeground - Background modeling and get foreground mask.

Calling Sequence

fg = detectforeground(im)

fg = detectforeground(im, method)

Parameters

• im: The input image.
• method: Background modeling methods. Can be 'LI'(default), 'GMM'. Other methods will appear in the future.
• fg: If detectforeground is used to create background model, then fg is an empty matrix. If detectforeground is used to detect foreground, then fg is the foreground mask.

Description

The function detectforeground creates background model and gets the foreground mask from an input video. The user should first use fg=detectforeground(im,method) or fg=detectforeground(im) to create the background model by the assigned method or the default method. Then use fg=detectforeground(im) to detect foreground and get the foreground mask.

fg=detectforeground(im, 'LI')

Create background model using the algorithm proposed by Liyuan Li etc. on ACM MM2003.

fg=detectforeground(im, 'GMM')

Create background model using the algorithm proposed by P. KadewTraKuPong and R. Bowden in 2001.

fg=detectforeground(im)

Create background model using the default algorthm('LI'). Or detect foreground and get the foreground mask;

Examples

 
    
    n = aviopen(SCI+'/contrib/sivp-svn-trunk/images/video.avi');
    im = avireadframe(n);
    fg = detectforeground(im, 'GMM');//or:fg = detectforeground(im). create background model
  
    while ~isempty(im),
          fg = detectforeground(im);//get the foreground mask
          imshow(fg);   
          im = avireadframe(n);
    end;

    aviclose(n);

Authors

Jia Wu <jiawu83[at]gmail.com>

Shiqi Yu <shiqi.yu[at]gmail.com>

Bibliography

'LI': "Foreground Object Detection from Videos Containing Complex Background", Liyuan Li, Weimin Huang, Irene Y.H. Gu, and Qi Tian, ACM MM2003.

'GMM': "An improved adaptive background mixture model for real-time tracking with shadow detection", P. KadewTraKuPong and R. Bowden, in Proc. 2nd European Workshp on Advanced Video-Based Surveillance Systems, 2001.

Availability

http://sivp.sourceforge.net

See Also