电子稳像技术算法研究

首先,本文介绍了电子稳像技术的基本原理,对2种常用的电子稳像技术算法进行了分析,并阐述了其优缺点.其次,本文提出了基于特征点的稳定视频图像序列电子稳像算法,其基本步骤是先获取图像特征点,然后确定补偿量,最后校正图像序列.利用基于特征点的匹配技术估计出全局运动矢量,用均值滤波的方法确定图像的补偿参数,并将补偿参数代入变换模型,继而实现了图像序列的稳定.实验结果表明,这种特征点稳像算法能够较好地稳定视频图像序列.

Research of electronic image stabilizing algorithm

Firstly, the basic principle of electronic image stabilizing technology was introduced. Two kinds of electronic image stabilizing algorithm were analyzed, whose merit and demerit were introduced in details. Secondly, an electronic image stabilization algorithm based on image features was proposed, and the basic step was detection of image features,estimation of image motion vector and ways of image composition were detailed. Globe motion vector between the reference frame and the current frames was obtained by a real-time electronic image stabilizing algorithm based on feature matching. The parameters to compensate for the current image can be determined by average value filtering and havebeen applied to give motion model to stabilize the image stream. The experiment results show that this algorithm can be used to achieve better real-time stabilization of image sequence.