论文部分内容阅读
提出了一种新的运动目标检测方法,这种方法可以有效的提取目标轮廓。应用一种图像差分技术得到运动目标的初始轮廓线。使用了动态轮廓线使其收敛到目标轮廓。提出了一种新的目标轮廓特征级融合方法,求解两类模式图像的收敛动态轮廓线控制点向量差的范数平方极小化。这种方法不需要图像配准降低了融合的计算复杂度,有效提高了可见光图像中目标轮廓提取的精度。对比检测实验证实了算法的有效性。设计了一种基于Newmark方法的动态轮廓线快速迭代算法,将该方法和方法作了比较,对比实验表明这种方法的时间复杂度降低了22%。
A new moving object detection method is proposed, which can effectively extract the target contour. Applying an Image Difference Technique to Obtain the Initial Contour of a Moving Object. Dynamic contour is used to converge to the target contour. A new target contour feature level fusion method is proposed to solve the norm square minimization of the vector difference of the control points of the convergent dynamic contour lines of two types of pattern images. This method does not require image registration to reduce the computational complexity of fusion and effectively improve the accuracy of target contour extraction in visible images. The contrast test verifies the effectiveness of the algorithm. A fast iteration algorithm of dynamic contour based on Newmark method is designed. The comparison between the method and the method shows that the time complexity of this method is reduced by 22%.