文章基本信息

标题：Adaptive Variable-size Search Window based on SURF
本地全文：下载
作者：Heba kandil ; Eman Eldaydamony ; Ahmed Atwan 等
期刊名称：International Journal of Multimedia and Ubiquitous Engineering
印刷版ISSN：1975-0080
出版年度：2013
卷号：8
期号：2
出版社：SERSC
摘要：Video tracking is a rich research point nowadays due to its wide range of applications such as surveillance. One of the challenges in video tracking is to exactly determine the location of the tracked object within each frame. Most of tracking algorithms make use of a fixed size search window regardless of the tracked object scale change over time. The fact is that too small search window may lose details of the tracked object. Besides, undue increase of computational complexity is resulted of inaccurate large search window. Adaptive variable-size search window algorithm is proposed to overcome these problems. Even if the tracked object is partially or completely occluded the algorithm should locate the expected location of it in an efficient way. The proposed algorithm is based on speeded up robust features (SURF). SURF is one of the fastest descriptors which generate a set of interest points that are invariant to various image deformations and robust against occlusion conditions during tracking. SURF points of the tracked object are extracted from the initially determined search window. The proposed algorithm makes use of the positional information of the extracted SURF points to update the size and location of the search window in the following frames. The results achieved more accuracy of the tracking process. The proposed algorithm produces a search window that is more fitted to the tracked object than search windows produced by common tracking algorithms such as mean shift do. Any tracking algorithm can make use of the proposed algorithm as it works in parallel with it to update the search window location and size to precisely track the object. Less computational time in the search window is an added value. Prediction of the exact location of the tracked object under occlusion condition is more precise than existing algorithms.
关键词：SURF; Mean shift; Search window; visual tracking