一种特征匹配的高精度电子稳像方法

High-accuracy electronic image stabilization method using feature matching

针对车载侦察系统视频稳像精度要求较高的情况,提出一种基于边缘图像配准的高精度电子稳像方法。该方法采用相位一致性作为不变度量来同时检测图像的角点和边缘特征;利用匹配点对引导边缘匹配,并筛选适量分布均匀的匹配点对求取初始全局进行运动估计;改进粒子滤波以对边缘图像进行配准来获得精确全局运动参数,进一步提高了稳像精度。实验结果表明,所提出方法的平移配准误差(Δx=+0.210,Δy=-0.013)远小于一个像素,满足了车载侦察系统对电子稳像技术的实时性及高精度要求。

In order to satisfy the high image stabilization precision requirement of reconnaissance system on vehicle,this paper presented a high-accuracy electronic image stabilization method based on edge image registration.The proposed method used phase congruency as invariant to detect corner and edge feature from the frame.Then utilized matched points to guide edge matching,and selected a proper number of well distributed corresponding points to get initial global motion estimation.Finally,to obtain accurate global motion parameters,applied improved particle filter to edge image registration to further improve the image stabilization precision.Experimental results show the translation registration errors of the method（Δx=＋0.210,Δy=-0.013） are far less than one pixel,thus the proposed method can satisfy the demand of real-time and high precision for electronic image stabilization technique in reconnaissance system on vehicle.