Active Contour Based Visual Tracking Using Level Sets

The visual tracking is the major process in finding the spot of moving object over time using a camera. Object tracking is challenging task when the object moves fast relative to the frame rate. The active contour algorithm is used for tracking the object in a given frame of an image sequence. In videos particular object motion can be tracked by using stationary cameras but in moving camera the particular object cannot be extracted from background subtraction. Active contour based visual tracking using level sets is proposed which does not consider the camera is stationary (or) moving .The optical flow based algorithm is used for initializing contours at first frame. The correlations between the values of neighbouring pixels for posterior probability estimation is done by Markov random field theory in the color based contour evolution. The Independent Component Analysis (ICA) is used to deal with noise (or) partial occlusion to obtain the more accurate contours in the shape based contour. To handle the abrupt motions the particle swarm optimization is used to track the object from first frame to last frame and it is applied to current frame to produce an initial contour in next frame. This visual tracking can be used in real time applications like vehicle guidance and control, surveillance and identification, user interface, video processing and medical applications.