电子稳像技术算法研究

摘要随着大规模集成电路的高速发展，电子稳像技术应用越来越广泛。对比于光学稳像和机电结合稳像，电子稳像更加完备，而更基础的电子稳像算法研究显得极其重要，对于进一步的应用是非常具有实际意义的。本文在分析常见的几种稳像方法的基础上，对电子稳像技术基本原理进行研究，分析了几种常见的电子稳像算法。其中，重点分析了基于特征点的电子稳像算法，其基本步骤是先得到图像的特征点，然后确定补偿量，最后对图像序列校正。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，并将补偿参数代人变换模型，继而实现了图像序列的稳定。实验表明，通过这种特征点稳像算法可以获得较好稳定效果。61572

毕业论文关键词：电子稳像；稳像算法；运动矢量估计

毕业设计说明书（论文）外文摘要

research of electronic image stablizing algorithm

Abstract With the rapid development of large-scale integrated circuits ,electronic image stabilization technology is applied more wildly .In contrast to the optical image stabilization and mechanical-electrical integration stabilization,electronic image is more perfect,and more basic electronic image stabilization algorithm is extremely important.It’s very practical significance for further application.Based on the analysis of several common image stabilization method,research the basic principles of electronic image stabilization technology research, analysis of several common electronic image stabilization algorithm. Which focus on analyzing the stability of feature points based on video sequences electronic image stabilization algorithm, the basic step is to get the image feature points, and then determine the compensation , the final corrected image sequences. The use of technology based on feature points matching the estimated global motion vector with mean filtering method to determine the parameters of the image compensation and compensation parameter generation transformation model, and then achieve a stable image sequence. Experimental results show that the feature point image stabilization algorithm can stabilize video image sequence.

Keyword:

electronic image stabilization;image stabilization algorithm;estimation of motion vector

1 绪论 6

2 稳像技术 7

2.1 稳像技术概述 7

2.2 视频产生模糊的机理 8

2.2.1 帧内模糊的产生机理 9

2.2.2 帧间模糊的产生机理 11

2.3 稳像技术分类 12

2.3.1 光学稳像 12

2.3.2 机械稳像 13

2.3.3 电子稳像 13

3 电子稳像技术研究 14

3.1 电子稳像技术的概念 14

3.2 电子稳像技术的基本原理 14

3.3 电子稳像技术的数学模型 15

3.4 电子稳像技术的特点 16

3.5 电子稳像的主要流程和核心环节 17

4 电子稳像算法研究 19

4.1 图像预处理 19

4.1.1 中值滤波电子稳像技术算法研究:http://www.youerw.com/tongxin/lunwen_67388.html