摘要
运动背景下的运动目标检测是一项难度大、计算量大的工程,它的难点是难以在实际工程中达到实时性的要求。本文针对这一问题提出了一种利用Lucas-Kanade和层次聚类相结合的方法,能够快速检测运动背景中的运动目标。实验证明该方法速度快,准确性高,在实际工程中具有一定的应用价值。
Moving object detection in the dynamic background is a hard complex task which has high time complexity.It is hard to meet its real time requirement in the project.In this paper,we present a new method that synthesize Lucas-Kanade algorithm and Hierarchical clustering to detect moving object rapidly.Finally,the experiments show that this method is efficient and accurate in the moving background.
出处
《国外电子测量技术》
2011年第12期44-47,共4页
Foreign Electronic Measurement Technology
关键词
运动目标
运动背景
光流检测
dynamic object
moving background
optical detect
