提出了一种基于Split Bregman迭代求解分段常值模型(也称为C-V模型)的运动目标检测方法。该方法首先采用高斯混合模型进行背景建模, 然后减去背景得到图像序列的运动区域部分(本方法的处理对象)。由于引入了Split Bregman迭代方案, 可以在保证演化过程稳定的前提下采用相对较大的时间步长。在真实红外序列图像上的检测情况表明, Split Bregman迭代方案加速了曲线的演化并且极大地降低了迭代过程所需的次数, 说明了该方法的有效性。

A moving target detection method was proposed based on the piecewise constant Mumford-Shah model (also known as the C-V model) solved by the Split Bregman iteration.In the method the Gaussian mixture model was used for background modeling and then the background was subtracted to obtain the moving regions of the image sequence (the handling objects of our method).As a result of the introduction of the Split Bregman iteration scheme we could use a rather large time step while maintaining the stability of the evolution process.The experimental results demonstrated that for real infrared image sequences the Split Bregman iteration scheme can accelerate the evolution of the curve and significantly reduce the number of iterations which proved the validity of our method.