Object Tracking in Video Images based on Image Segmentation and Pattern Matching

Abstract

The moving object tracking in video pictures [1] has attracted a great deal of interest in computer vision. For object recognition, navigation systems and surveillance systems [10], object tracking is an indispensable first-step. We propose a novel algorithm for object tracking in video pictures, based on image segmentation and pattern matching [1]. With the image segmentation, we can detect all objects in images no matter whether they are moving or not. Using image segmentation results of successive frames, we exploit pattern matching in a simple feature space for tracking of the objects. Consequently, the proposed algorithm can be applied to multiple moving and still objects even in the case of a moving camera. We describe the algorithm in detail and perform simulation experiments on object tracking which verify the tracking algorithm‘s efficiency. VLSI implementation of the proposed algorithm is possible. The conventional approach to object tracking is based on the difference between the current image and the background image. However, algorithms based on the difference image cannot simultaneously detect still objects. Furthermore, they cannot be applied to the case of a moving camera. Algorithms including the camera motion information have been proposed previously, but, they still contain problems in separating the information from the background. The proposed algorithm, consisting of four stages i.e. image segmentation, feature extraction as well as object tracking and motion vector determination [12]. Here Image Segmentation is done in 3 ways and the efficiency of the tracking is compared in these three ways, the segmentation techniques used are ―Fuzzy C means clustering using Particle Swarm Optimization [5],[6],[17]”, ”Otsu’s global thresholding [16]”, ”Histogram based thresholding by manual threshold selection”, after image segmentation the features of each object are taken and Pattern Matching [10],[11],[20] algorithm is run on consecutive frames of video sequence, so that the pattern of extracted features is matched in the next frame , the motion of the object from reference frame to present frame is calculated in both X and Y directions, the mask is moved in the image accordingly, hence the moving object in the video sequences will be tracked.