适用于扑翼飞行器的双滤波电子稳像方法

Dual-Filter Electronic Image Stabilization Method Suitable for Flapping-Wing Aircraft

针对扑翼飞行器航拍视频抖动问题,提出一种基于角点检测的双滤波电子稳像算法。首先使用FAST角点检测算法和KLT算法,求解图像内特征点的局部运动矢量,并通过最小二乘法拟合得到帧间的全局运动矢量,然后采用Kalman滤波和Savitzky-Golay滤波相结合的双滤波算法,从全局运动参数中提取出需要保留的主动运动参数,之后利用相邻帧补偿方法对图像进行运动补偿,接着通过裁剪法修复未定义区,最终得到相对稳定的视频。实验结果表明,上述方法能够有效减小扑翼飞行器航拍视频的抖动现象。

Aiming at the problem of flapping-wing aircraft aerial video jtter,a dual-filter electronic image stabilization algorithm based on corner detection is proposed.First,the FAST corner detection algorithm and the KLT algorithm were used to solve the local motion vector of the feature points in the image,and the Least Squares Method was used to solve the global motion vector between adjacent frames.Then,the active motion parameters that need to be preserved were extracted from the global motion parameters using the double filtering algorithm combining Kalman filter and Savitzky-Golay filter;the motion compensation method of adjacent frames was used to compensate the image.After that,the undefined area was repaired by clipping method,and a relatively stable video was finally obtained The experimental results show that the above method can effectively reduce the shaking of the flapping-wing aircraft aerial video phenomenon.