基于光流模型的图像运动估计方法

Method of estimating image motion based on the optical flow model

运动图像序列参数估计是电子稳像的关键技术之一。安装在运动载体上的摄像机采集图像时 ,由于载体姿态变化产生不平稳运动 ,这种不平稳运动将影响摄像系统的成像质量。电子稳像的目的就是通过计算运动参数并校正图像 ,以保证处理后的图像序列达到稳定效果。光流模型是处理运动图像的有效方法 ,本文介绍了用光流模型计算运动图像参数的基本原理和步骤。根据图像像素强度守恒原理 ,建立光流约束方程 ,计算运动参数 ,最后结合实例计算两帧样本图像之间水平和垂直位移量 ,并绘制光流矢量图。

It is one of the important technologies in electronic stabilizing images to estimate the sequence of image motion. A camera mounted on a moving vehicle collects a sequence of images. We consider a vehicle's motion to be non-smooth when impulsion or high-frequency changes occur, the non-smooth motion will perturb the sequence of images obtained by the camera. The goal of electronic stabilizing images is to estimate the motion quantities, and correct the sequence of images, so that they are approximately the same as the images that would have been obtained if the motion of the vehicle had been smooth. Using the optical flow model to process the image motion is a good method. This paper presents the basic theory and approach to calculating image motion by using the optical flow model. It shows how to derive the optical flow constraint equation, and how to use it to estimate motion quantities. Finally, It gives the vertical and horizontal displacement between the consecutive two frames and the map of optical flow vectors.