基于曲线进化的目标检测算法

提出了基于曲线进化的运动目标检测算法。对于运动目标检测问题定义了新颖的能量模型,并给出了相应的曲线进化方程。结合水平集算法,所提出的曲线进化模型能够自动的适应拓扑变化,检测出多个运动目标区域及轮廓。从计算物理学领域引入基于偏微分方程的水平集窄带算法,在构造窄带时无需显示的知道曲线的位置,并能够快速的实现。

Moving Object Detection Algorithm Based on Curve Evolution

This paper proposes an algorithm for moving object detection based on curve evolution theory. A novel energy model for object detection is defined and the corresponding curve evolution equation is given. Combined with level set method, the proposed model can be adapted to topological change automatically, and detect multiple moving targets and their contours. A PDE-based local fast level set method is introduced that originated from computational physics. With this new algorithm used, there is no need to know the position of curve explicitly when constructing narrow bands. This newly introduced algorithm can also be implemented quickly.