摘要
SIFT算法是特征图像特征提取中一种最具鲁棒性的算法,但是其在特征提取匹配上速度较慢,很难满足实时目标跟踪的要求。使用SURF特征提取方法既保持了SIFT算法的高精度的优点,又克服了速度慢的缺陷。提出使用SURF提取并且匹配目标的特征点,用重心算法计算目标的脱靶量,通过小区域跟踪方法和高速硬件平台实现目标的实时跟踪。实验证明,算法对目标的轻微旋转、部分遮挡、亮度变化具有很强的鲁棒性,跟踪速度比SIFT算法也极大提高。
SIFT is a kind of the most robust algorithm in image feature extracting,but feature extracting and matching speed is slow which can't meet the requirement of real time target tracking.The proposed method uses SURF feature extracting method which has the advantage of high precision of SIFT and overcome the disadvantage of low speed.First,it uses SURF algorithm to extract target feature and match them in following frames.Then,the bary method is used to compute the target displacement.The proposed method is programmed in high speed hardware system and using small zone tracking method in order to meet the real time requirement.Experimental results show that this method has strong robustness to little rotated, shielded,illumination changed of the target and decrease largely the computing time compared with SIFT.
出处
《长春理工大学学报(自然科学版)》
2011年第2期138-141,153,共5页
Journal of Changchun University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助(60873163)