期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2015
期号：MULTICON
页码：494
出版社：S&S Publications
摘要：Object Tracking is used to associate target objects in consecutive video frames. The association can beespecially difficult when the objects are moving fast relative to the frame rate. This paper deals with object trackingfrom videos based on template matching using gaussian kernel with probability distribution function and mean shiftalgorithm. Initially, the target will be selected from chosen video sequence to track desired object in consecutiveframes. The target will be utilized to determine the probability distribution function for similarity measurementbetween target and current processing frames. Here, gaussian kernel function and its gradient are used to find the PDFfor corresponding templates. Similarity between two different images will be measured by weighted sum of gaussiancoefficients and PDFs. Mean shift approach used here to shift the starting coordinates of template to find its similarfeatures in consecutive frames to detect the desired object. The dissimilarity between the target model and targetcandidates will be expressed by a metric derived from Bhattacharyya coefficient. The project simulated results showsthat moving object from video will be tracked accurately at different position and shape with help of templates in aconsiderable amount of time.
关键词：Bhattacharyya coefficient; Gaussian Coefficient; Mean Shift algorithm; Template Modelling
