一种基于光流的无人机序列影像背景运动补偿方法

Background Motion Compensation Method with Unmanned Aerial Vehicle Image Sequence Based on Optical Flow

针对无人机序列影像,采用一种鲁棒的变分光流模型求解无人机序列影像光流场,该模型采用亮度守恒和梯度守恒结合的数据项,平滑项则采用二阶的Laplacian算子约束项,结合多分辨率影像技术,采用由粗到精的策略进行优化。利用光流场估计的影像运动信息对无人机序列影像进行背景运动补偿。实验结果表明,该光流模型获得的光流场能够准确的估计出无人机序列影像背景的运动,并取得了较好的背景运动补偿效果。

A robust variable optical flow model is used to solve the optical flow of UAV image se- quences. This model consists of data terms combining the brightness constancy assumption with the gradient constancy assumption, and smoothness terms using Laplacian. Combined with multi-resolution image technology, a coarse-to-fine strategy is used to optimize the model. The background motion is compensated by the optical flow information. Experimental results show that the optical flow obtained by this model can accurately evaluate the background motion, and achieves good compensation.