由形态学边缘模式匹配实现数字稳像

王小鹏; 阎国梁; 裴建刚; 晏鑫

doi:null

您当前的位置：

首页 >

文章列表页 >

由形态学边缘模式匹配实现数字稳像

信息科学 | 更新时间：2020-08-12

- 由形态学边缘模式匹配实现数字稳像
- Stabilization of digital image by morphological edge pattern matching
- 光学精密工程 2009年17卷第3期页码：683-689
- 作者机构：
  
  兰州交通大学电子与信息工程学院,甘肃兰州,730070
- 作者简介：
- 基金信息：
  
  甘肃省自然科学基金资助项目(No.0710RJZA046)();兰州交通大学"();青蓝"();人才工程基金资助项目()
- DOI：
  中图分类号： TN941.1;TP391.4
- 收稿日期：2008-07-08，
  
  修回日期：2008-07-25，
  
  网络出版日期：2009-03-25，
  
  纸质出版日期：2009-03-25
- 稿件说明：
移动端阅览
王小鹏, 阎国梁, 裴建刚, 晏鑫. 由形态学边缘模式匹配实现数字稳像[J]. 光学精密工程, 2009,17(3): 683-689

WANG Xiao-peng, YAN Guo-liang, PEI Jian-gang, YAN Xin. Stabilization of digital image by morphological edge pattern matching[J]. Editorial Office of Optics and Precision Engineering, 2009,17(3): 683-689
王小鹏, 阎国梁, 裴建刚, 晏鑫. 由形态学边缘模式匹配实现数字稳像[J]. 光学精密工程, 2009,17(3): 683-689 DOI：

WANG Xiao-peng, YAN Guo-liang, PEI Jian-gang, YAN Xin. Stabilization of digital image by morphological edge pattern matching[J]. Editorial Office of Optics and Precision Engineering, 2009,17(3): 683-689 DOI：

摘要

针对在一些实际应用场合摄像机设备的震动造成的视频序列失稳

提出了一种基于形态学边缘模式匹配的数字稳像方法。该方法首先利用形态学方法提取视频图像边缘

然后利用当前帧子块与参考帧子块的边缘进行特征匹配

以确定当前帧子块相对于参考帧的局部运动矢量。对得到的局部运动矢量进行分析判决

以确定该局部运动矢量是否为真正的抖动偏移量

消除视频帧本身因存在局部运动目标、低对比度、纹理区域等因素对全局运动矢量估计的影响;最后

通过判别确认的局部运动矢量确定全局运动矢量并进行运动补偿

实现数字稳像。仿真实验表明

由于该方法的块局部运动矢量采用1 bit边缘进行最大相关值估计

因而复杂度低

而采用异常情况判决准则消除了其对真正运动矢量估计的影响

稳像精度高

可应用于因摄像头平移震动等因素造成的视频序列失稳。

Abstract

To eliminate the image jittering from the vibration caused by the translation motion of a camera

a method of digital image stabilization based on morphological edge pattern matching is proposed. In this method

morphological edge matching is used to estimate the local movement vectors based on the 1 bit edge maximal correlation. Since correlation calculation is logical operator instead of pixel-wise gray arithmetic operators

the complexity of proposed method is greatly reduced. Moreover

in order to remove the undesired local motion vectors caused by several undesired conditions

the local motion vectors are analyzed to judge whether it is desired or not and to eliminate the effects of locate motion target

low contrast and texture pattern regions on the estimation of globe motion vectors.Finally

the reliable globe movement vector is obtained by weight median. Experiments show that this method is fast and efficient for image stabilization

and can be applied in the video vibrations without large rotations.

关键词

Keywords

references

HSU S C,LIANG S F,LIN C T. A robust digital image stabilization technique based on inverse triangle method and background detection[J]. IEEE Transactions on Consumer Electronics, 2005,51(2):335-345.[2] 赵红颖,金宏,熊经武. 电子稳像技术概述[J]. 光学精密工程,2001,9(4):353-359. ZHAO H Y,JIN H,XIONG J W.Overview of the electronic image stabilization technology[J].Opt. Precision Eng.,2001,9(4):353-359.(in Chinese)[3] 孙辉. 快速灰度投影算法及其在电子稳像中的应用 . 光学精密工程,2007,15(3):412-416. SUN H. Fast gray projection algorithm and its application[J].Opt. Precision Eng., 2007,15(3):412-416. (in Chinese)[4] HSU S C,LIANG S F,FAN K W,et al.. A robust in-car digital image stabilization technique [J]. IEEE Trans. On Systems, Man and Cybernetics, 2007,37(2):234-247.[5] BOMBINI L,CERRI P,GRISLERI P,et al.. An evaluation of monocular image stabilization algorithms for automotive applications . Proceedings of the IEEE ITSC 2006, Conference IEEE Intelligent Transportation Systems, Toronto, Canada, 2006:1562-1567.[6] KO S J,LEES H,JEON S W. Fast digital image stabilizer based on gray-coded bit-plane matching [J]. IEEE Trans. Consumer Electronics, 1999,45(3):598-603.[7] PARIK J K,PARK Y C,KIM D W. An adaptive motion decision system for digital image stabilizer based on edge pattern matching [J]. IEEE Trans. On Consumer Electronics,1992,38(3):607-616.[8] ZHANG Y J. Image Engineer (II)-Image Analysis (Second Edition) [M]. Beijing:Tsinghua University, 2005. (in Chinese)[9] KO S J,LEE S H,LEE K H. Digital image otabilization based on bit-plane matching [J]. IEEE Transactions on Consumer Electronics, 1998,44(3):617-622.

浏览量

527

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据