Moving Object Recognition and Tracking Supat Wongwirawat and Metha Jeeradit ee 547 Computer Vision Abstract



Download 139.94 Kb.
Page2/3
Date01.06.2018
Size139.94 Kb.
#52665
1   2   3

Start


Read first 2 image frames





Temporal differencing.

M(i,j) = F2(i,j) , | F2(i,j)-F1(i,j) | >= Threshold

= 0 , | F2(i,j)-F1(i,j) | < Threshold

where F2 is the current image, F1 is the previous image, and M is the motion image.






First Segment

Segments each motion region using recursive function setlabel( ), and get rid of any region that is too small. (size < prespecified object size) This step gives the approximated sketch of the moving object. We will use this sketch to segment the full object.






Download 139.94 Kb.

Share with your friends:
1   2   3




The database is protected by copyright ©ininet.org 2024
send message

    Main page