07-11-2014, 10:44 AM
Abstracts: Object tracking in video sequence is a very important topic and has various applications in surveillance, robot technology, video compression, etc. In many applications the main goal is on tracking moving objects. In this paper we will track multiple vehicles in fix camera with specific shape and to perform this operation we can use Kalman filter, Optical flow and Gaussian mixture model and also count multiple objects in video frame. Each time we detect an object and that detected object does not correspond to a previously existing object, we recognize that object as being either car or bike and increment the counter for the respective type. Optical flow is the pattern of apparent motion of objects in a video sequence. The optical flow method try to calculate the motion between two image frames which are taken at times t and t+ dt at every position. Here the method used in optical flow is horn-schunk method. The advantages of Optical Flow are quick calculations and the disadvantage is a lack of complete object tracking and it is very sensitive to the noise. The Kalman filter is an efficient recursive filter which estimates the state of a dynamic system from a series of incomplete and noisy measurements. This method can provide accurate continuously updated information about the position and velocity of vehicle. GMM can be used in the context of a complex environment. The Expectation-Maximization (EM) algorithm, is a method of determining a best fit GMM. The advantage of GMM is complete results of operation the disadvantage is not a complete object tracking