摘要
运动目标图像的检测跟踪技术是计算机视觉的关键技术之一,广泛应用于工业、交通、军事等诸多领域。基于无人机平台下的动态场景运动目标检测与跟踪是该领域的一个技术难点,在实际应用中,环境复杂、目标较小和平台运动等特性对算法的实时性和可靠性提出了较高的要求。如果能将目标的一些其它特征加入到搜索过程中就有可能改善Camshift本身存在的不足。本文以跟踪算法Camshift为基础,将SURF特征检测整合到Camshift算法中,大大提高了目标物体的跟踪准确度,同时也有较好的实时性,能够对目标实现更好的跟踪,实现了较好的跟踪效果。
The detection and tracking technology of moving image is one of the key technologies in computer vision area,which is widely used in industry,transportation,military and other fields.Dynamic scene moving target detection and tracking based on UAV platform is a technical difficulty,especially in practical application. The features such as complex environment,small target and platform motion have higher requirements on the real-time performance and reliability of the algorithm. There is a possibility to improve Camshift itself by add some other features of the target to the search process. This paper will be integrated into the Camshift algorithm with SURF feature detection,it is not only improving the target tracking accuracy of the object,but also has a good real-time performance,It can achieve better tracking effect.
作者
樊伊君
梁朝钢
FAN Yi-jun;LIANG Chao-gang(No.1 Institute of Geological ＆ Mineral Resources Survey of Henan,Luoyang 471023,China)
出处
《电子设计工程》
2018年第10期105-108,共4页
Electronic Design Engineering
基金
中国地勘基金(1212011220503)
