|本期目录/Table of Contents|

[1]姜哲颖,周华兵*,刘 姣.SIFT特征匹配和VFC算法的电子稳像技术[J].武汉工程大学学报,2015,37(09):45-49.[doi:10. 3969/j. issn. 1674-2869. 2015. 09. 008]
 IANG,HOU,IU.Electronic Image Stabilization Based on SIFT Matching and VFC[J].Journal of Wuhan Institute of Technology,2015,37(09):45-49.[doi:10. 3969/j. issn. 1674-2869. 2015. 09. 008]
点击复制

SIFT特征匹配和VFC算法的电子稳像技术(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
37
期数:
2015年09期
页码:
45-49
栏目:
机电与信息工程
出版日期:
2015-09-30

文章信息/Info

Title:
Electronic Image Stabilization Based on SIFT Matching and VFC
文章编号:
1674-2869(2015)09-0045-05
作者:
姜哲颖周华兵*刘 姣
武汉工程大学计算机科学与工程学院,湖北 武汉 430205
Author(s):
JIANG Zhe-yingZHOU Hua-bingLIU Jiao
School of Computer Science and Engineering,Wuhan Institute of Technology,Wuhan 430205,China
关键词:
运动估计向量场一致性电子稳像
Keywords:
motion estimationvector field consensuselectronic image stabilization
分类号:
TP391
DOI:
10. 3969/j. issn. 1674-2869. 2015. 09. 008
文献标志码:
A
摘要:
为了处理较为模糊的抖动视频,提出了一种基于尺度不变特征转换(SIFT)的特征匹配和向量场一致(VFC)优化算法的电子稳像技术. 该技术着重于研究视频的运动估计阶段,利用SIFT提取高独特性的特征点,根据特征点进行帧间匹配,结合VFC,通过向量场的学习将外点从内点区分开来,得到两帧图像中特征序列的平移轨迹,然后估算出运动补偿向量并校正每一帧图像的相对位置,从而输出稳定视频. 实验表明,该电子稳像技术处理抖动视频时精度高和耗时短.
Abstract:
To process the blurry shaky video, electronic image stabilization based on scale invariant feature transform (SIFT) matching and vector field consensus (VFC) was proposed, which is mainly applied to video motion estimation. First, SIFT was used for extracting highly distinctive invariant feature and setting up the initial feature matching. Then, by the vector field consensus algorithm, we distinguished inliers from outliers and obtained the characteristics of translational trajectory in sequence of two video frames. Finally, we estimated the motion compensation to correct the relative position of every video frame by trajectory. Experiment results show the advantages of the method in precision and efficiency.

参考文献/References:

[1] 张跃飞.车载摄像机数字稳像技术研究[M].成都:电子科技大学,2011.ZHANG Yue-fei.The research of the digital image stabilization technology for the In?鄄Car cameras[M].Chengdu: University of Electronic Science and Technology of China,2011.(in Chinese)[2] KUMAR S, AZARTASH H, BISWAS M, et al. Real-time affine global motion estimation using phase correlation and its application for digital image stabilization[J]. IEEE Trans Image Process,2011,20(12):3406-3418.[3] OKADE M, BISWAS P K. Video stabilization using maximally stable extremal region features[J]. Multimedia Tools Appl,2014, 68( 3): 947-968.[4] 王海晖,卢培磊,吴云韬,等. 无参考视频平滑度的评价方法[J]. 武汉工程大学学报,2015,37(6):56-62. WANG Hai-hui,LU Pei-lei,WU Yun-tao,et al. Evaluation method of no-reference video smoothness[J].Journal of Wuhan Institute of Technology,2015,37(6):56-62.(in Chinese) [5] BATTIATO S, GALLO G, PUGLISI G,et al. SIFT features tracking for video stabilization[J]. IEEE Computer Society, 2007,27(5):825-830.[6] ZHAO Ji, MA Jiayi, JIN Wen et al. A robust method for vector field learning with application to mismatch removing[J]. Computer Vision and Pattern Recognition (CVPR)2011 IEEE Conference on,2011,30(6):2977-2984.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2015-08-28基金项目:湖北省教育厅科学技术研究项目(Q20151503)作者简介:姜哲颖(1991-),男,湖北武汉人,硕士研究生.研究方向:数字图像处理.* 通信联系人.
更新日期/Last Update: 2015-10-19