噪声条件下基于粒子群优化的数字稳像方法

Digital image stabilization based on particle swarm optimization under noisy condition

当视频序列中同时存在随机噪声和随机晃动时,传统的数字稳像算法由于受到噪声干扰而无法有效消除视频序列中的随机晃动。为了稳定这种存在随机噪声的视频序列,提出了一种基于粒子群优化的数字稳像方法。首先,定义了衡量寻优结果适应度函数,即输入视频连续若干帧均值图像的能量;然后,算法利用粒子群优化策略来搜索视频序列的最优运动补偿向量;最后,实验分别使用模拟抖动视频和真实拍摄的视频来测试算法的性能。实验结果表明,当测试视频同时存在随机噪声和随机晃动时,该算法不仅能够有效消除视频的随机晃动,并且有效抑制了随机噪声。

The classical digital image stabilization methods are unable to remove the undesired frame jitters of unstable videos,when the videos not only affected by frame jitters but also by random noise.This paper proposed a digital image stabilization method based on particle swarm optimization for noisy videos.Firstly,defined the fitness function evaluating the optimization result which was generated by averaging a series of consecutive frames of the input video.Then,utilized the particle swarm optimization to search the optimal motion compensating vectors.In the end,used the simulated unstable video and real one to evaluate the performance of the proposed method.The experimental results show that the proposed method can not only remove the random jitters but also reduce the random noise.