Abstract

The object representation and tracking is one of the important tasks in computer vision. The object can be represented in various ways and in this paper the objects are represented using the properties of the HSV color space. Adaptive k-means clustering algorithm was applied to cluster objects centroids color values and co-ordinates were sent to next frame for clustering. After clustering, for comparing the objects present in both the reference frame and the target frame, a similarity measure was proposed which uses position, color and size of the objects for comparison. Based on the similarity value, the objects were detected and tracked. The performance of the proposed approach was verified with human objects and the same was effectively tracked. The comparison was carried with similar methods and the results are encouraging.