本文以图像内容分割为基础,提出了一种飞行目标运动估计新算法。该方法首先对机载视频图像进行预处理,并在HSV空间将图像内容划分为目标和背景两大类;然后根据使用者对目标和背景的不同质量要求,采用累积法对飞行目标的运动轨迹进行二次曲线拟合。试验表明,该方法可以提高运动估计速度,而且精度不亚于经典的运动估计算法。 Based on the segmentation of image content, this paper proposes a new algorithm for the motion estimation of flight targets. The method firstly preprocesses the on-board video images and divides the image content into two categories: target and background in the HSV space. Then, according to different quality requirements of the user on the target and the background, the cumulative trajectory of the moving target Perform quadratic curve fitting. Experiments show that this method can improve the speed of motion estimation, and its accuracy is as good as the classical motion estimation algorithm.