旋转视频中特征点的迭代筛选与光流估计匹配研究

doi:10.3969/j.issn.1000-1093.2012.11.009

兵工学报 ›› 2012, Vol. 33 ›› Issue (11): 1329-1334.doi: 10.3969/j.issn.1000-1093.2012.11.009

旋转视频中特征点的迭代筛选与光流估计匹配研究

王斌锐，徐崟，金英连，吴善强

（中国计量学院机电工程学院，浙江杭州 310018）

收稿日期:2011-03-31 修回日期:2011-03-31 上线日期:2014-01-10
作者简介:王斌锐（1978—），男，副教授
基金资助:
国家自然科学基金项目（50905170）；浙江省自然科学基金项目（Y1090042）

Iterative Filter and Optical Flow Estimation Matching of Feature Points in Rotary Jitter Compensation

WANG Bin-rui， XU Yin， JIN Ying-lian， WU Shan-qiang

(College of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018，Zhejiang, China)

Received:2011-03-31 Revised:2011-03-31 Online:2014-01-10

摘要/Abstract

摘要： 转动抖动补偿是视频稳像的难点，针对转动抖动补偿中的关键技术特征点的筛选与匹配展开研究。建立了图像的6参数仿射模型；推导得到估计有意运动参数的超定方程；采用最小二乘迭代算法来去除绝对误差和(SAD) 算法误判的特征点；采用金字塔(LK)光流算法来对旋转视频进行特征点匹配。编程实现算法；用特征窗口梯度矩阵法(KLT)提取特征后，分别用SAD算法和LK光流算法进行匹配，求解得到旋转变换阵参数误差，分析、比较并图示了误差原因；利用Kalman滤波去除无意运动；对含转动抖动的视频进行稳像补偿。在自主移动机器人平台上开展了实验。结果表明LK光流算法相比SAD算法对旋转视频的特征点匹配误差小，结合Kalman滤波可有效补偿转动抖动，将最大8.37°的转动抖动稳像到3.68°以下。

关键词: 信息处理技术, 特征点, 光流, 抖动补偿, 机器人

Abstract: Rotary jitter compensation is a difficulty in video image stabilization. The feature point matching and inaccurate point filtering were studied. An affine model with 6 parameters was established for moving images, and an over-determined equation to estimate motion parameters was derived. The least squares iterative algorithm was used to remove error feature points judged by sum of absolute difference (SAD) matching algorithm. A pyramid-style Lucas-Kanade (LK) algorithm based on optical flow was adopted for feature point matching of rotary video. All algorithms were programmed. After detecting feature points by using the gradient matrix of the feature window (KLT) method, SAD and LK algorithms were used to match feature points respectively, and the rotation matrix parameter errors were obtained and compared. The reasons of matching error were analyzed. Kalman filter was used to smooth rotation parameters. All algorithms were implemented on an autonomous robot. The experiment results show that LK can get less matching errors than SAD for rotation jitter, and Kalman filter makes the maximum 8.37° rotation jitter less than 3.68°, thus it can compensate the rotary video effectively.

Key words: information processing, feature point, optical flow, jitter compensation, robot

中图分类号:

TP242

王斌锐，徐崟，金英连，吴善强. 旋转视频中特征点的迭代筛选与光流估计匹配研究[J]. 兵工学报, 2012, 33(11): 1329-1334.

WANG Bin-rui， XU Yin， JIN Ying-lian， WU Shan-qiang. Iterative Filter and Optical Flow Estimation Matching of Feature Points in Rotary Jitter Compensation[J]. Acta Armamentarii, 2012, 33(11): 1329-1334.

参考文献

［1］孟龙，林行刚，王贵锦，等. 视频抖动矫正系统中的运动滤波[J] . 清华大学学报，2005, 45(1):41-43.
MENG Long, LIN Xing-gang，WANG Gui-jin, et al. Motion filter for video stabilizing systems[J]. Jounal of Tsinghua University,

2005, 45(1):41-43. (in Chinese)
[2] Onishi H, Suzuki H, Chamberlain S. Detection of rotation and parallel translation using hough and Fourier transforms [J]. IEEE Transaction on Image Processing, 1996, 3: 827-830.
[3] 尹德森，赵跃进. 基于角点跟踪的数字稳像算法[J]. 红外与激光工程, 2008,37(5):920-923.
YIN De-sen, ZHAO Yue-jin. Digital image stabilization algorithm based on corner tracking[J]. Infrared and Laser Engineering, 2008, 37(5):920-923. (in Chinese)
[4] Battiato S, Gallo G, Puglisi G. SIFT features tracking for video stabilization[C]∥14th International Conference on Image Analysis and Processing. Modena: IEEE, 2007: 825-830.
[5] Chung C Y, Chen H H. Feature-based full-frame image stabilization[C]∥Ninth IEEE International Symposium on Multimedia. Taiehung: IEEE, 2007:100-106.
[6] Liang Y M, Tyan H R, Chang Shyang-lih, et al. Video stabilization for a camcorder mounted on a moving vehicle[J]. IEEE Transactions on Vehicular Technology, 2004, 53(6): 1636-1648.
[7] 俞能海，陈豪，刘政凯. 基于块金字塔的快速块匹配算法[J].电路与系统学报, 2003, 8(3):54-57.
YU Neng-hai, CHEN Hao, LIU Zheng-kai. A fast block matching algorithm based on block sum pyramid[J]. Journal of Circuits and Systems,2003, 8(3):54-57. (in Chinese)
[8] Essannouni F, Thami R O H, Salam A, et al. A new fast full search block matching algorithm using frequency domain [C]∥The Eighth International Symposium on Signal Processing and Its Applications. Sydney: IEEE, 2005: 559-562.
[9] Cai J H, Walker R. Robust motion estimation for camcorders mounted in mobile platforms[C]∥The Conference of the Australian Pattern Recognition Society on Digital Image Computing: Techniques and Applications. Canberra: IEEE, 2008:491-496.
[10] 徐崟,王斌锐,金英连. 基于特征匹配和卡尔曼滤波的机器人视觉稳像[J]. 计算机工程，2011, 37（20）:194-196, 199.
XU Yin, WANG Bin-rui, JIN Ying-lian. Robot vision image stabilization based on feature matching and Kalman filtering
[J]. Computer Engineering, 2011, 37(20): 194-196, 199. (in Chinese)

[1]	刘江涛, 周乐来, 李贻斌. 复杂地形六轮独立驱动与转向机器人轨迹跟踪与避障控制[J]. 兵工学报, 2024, 45(1): 166-183.
[2]	岳胜哲, 王正杰. 基于实例分割与光流的动态环境SLAM[J]. 兵工学报, 2024, 45(1): 156-165.
[3]	许鹏, 邢伯阳, 刘宇飞, 李泳耀, 曾怡, 郑冬冬. 基于扩张状态观测器和模型预测方法的四足机器人抗干扰复合控制[J]. 兵工学报, 2023, 44(S2): 12-21.
[4]	刘思宇, 廖峻北, 雷飞, 王志瑞, 闫曈, 党睿娜, 郭朝. 用于四足机器人的并联弹性腿足关节设计与优化[J]. 兵工学报, 2023, 44(S2): 71-83.
[5]	降晨星, 姚其昌, 许鹏, 周雨婷, 闫曈. 新技术形势下四足、双足机器人技术的变革[J]. 兵工学报, 2023, 44(S2): 84-89.
[6]	鞠爽, 王晶, 王灏, 周萌. 基于GWO-WOA的执行器严重故障多移动机器人编队重构控制[J]. 兵工学报, 2023, 44(S2): 114-125.
[7]	许鹏, 赵建新, 范文慧, 邱天奇, 江磊, 梁振杰, 刘宇飞. 四足机器人特定复杂运动技能控制[J]. 兵工学报, 2023, 44(S2): 135-145.
[8]	李曹妍, 郭振川, 郑冬冬, 魏延岭. 基于分布式模型预测控制的多机器人协同编队[J]. 兵工学报, 2023, 44(S2): 178-190.
[9]	许威, 苏波, 江磊, 闫曈, 许鹏, 王志瑞, 邱天奇. 足式越野机器人集群系统关键技术与应用展望[J]. 兵工学报, 2023, 44(9): 2568-2579.
[10]	苏波, 江磊, 刘宇飞, 邢伯阳, 李泳耀, 谭森起, 王志瑞. 移动机器人跨域跃质关键技术综述[J]. 兵工学报, 2023, 44(9): 2556-2567.
[11]	李姗姗, 王洪波, 陈建宇, 张兴超, 田俊杰, 牛建业. 新型四足并联军用机器人步态控制算法及仿真[J]. 兵工学报, 2023, 44(3): 895-909.
[12]	尚哲, 王挺, 徐瑶, 吴英彪, 邵沛瑶, 邵士亮. 六轮摇臂移动机器人结构设计与越障动力学研究[J]. 兵工学报, 2023, 44(11): 3478-3488.
[13]	李辉, 宁峰平, 郭辉, 李瑞琴. 3-PRPS/RRR踝关节康复机器人运动机理研究与试验[J]. 兵工学报, 2023, 44(10): 3026-3037.
[14]	姜祎, 王挺, 邵沛瑶, 徐瑶, 邵士亮. 一种轮腿复合型机器人的步态研究与越障性能分析[J]. 兵工学报, 2023, 44(1): 247-259.
[15]	宋纪元, 朱爱斌, 屠尧, 张甲林, 张育林. 单兵救援助力外骨骼机器人的设计与特性分析[J]. 兵工学报, 2022, 43(9): 2037-2047.

旋转视频中特征点的迭代筛选与光流估计匹配研究

Iterative Filter and Optical Flow Estimation Matching of Feature Points in Rotary Jitter Compensation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价