基于核回归修正的上采样相位相关精确运动估计算法

doi:10.11772/j.issn.1001-9081.2016.08.2316

计算机应用 ›› 2016, Vol. 36 ›› Issue (8): 2316-2321.DOI: 10.11772/j.issn.1001-9081.2016.08.2316

基于核回归修正的上采样相位相关精确运动估计算法

余应淮, 谢仕义, 梅其祥

广东海洋大学信息学院, 广东湛江 524088

收稿日期:2016-01-13 修回日期:2016-03-11 出版日期:2016-08-10 发布日期:2016-08-10
通讯作者: 谢仕义
作者简介:余应淮(1981-),男,广东汕头人,硕士,主要研究方向:图像处理、模式识别、计算机视觉;谢仕义(1963-),男,四川巴中人,教授,硕士,主要研究方向:数字城市、海洋遥感、图像处理;梅其祥(1973-),男,湖南常德人,副教授,博士,主要研究方向:信息安全、智能信息处理。
基金资助:
国家自然科学基金资助项目（61272534）；广东海洋大学创新强校工程项目（2015KQNCX056）；湛江市科技计划项目（2015B01009）。

Accurate motion estimation algorithm based on upsampled phase correlation with kernel regression refining

YU Yinghuai, XIE Shiyi, MEI Qixiang

College of Information, Guangdong Ocean University, Zhanjiang Guangdong 524088, China

Received:2016-01-13 Revised:2016-03-11 Online:2016-08-10 Published:2016-08-10
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61272534), the Project of Enhancing School with Innovation of Guangdong Ocean University (2015KQNCX056), the Science and Technology Program of Zhanjiang (2015B01009).

摘要/Abstract

摘要： 针对亚像素运动矢量的精确估计问题，提出一种基于核回归修正的上采样相位相关精确运动估计算法。首先，使用矩阵相乘离散傅里叶变换方法快速计算上采样相位相关曲面，并通过检测其峰值坐标实现运动矢量的亚像素级初始估计；其次，在上采样相位相关曲面上，采用核回归方法对以初始估计值为中心的邻域进行拟合；最后，检测核回归拟合函数的峰值坐标，并以此坐标对初始估计值进行修正，从而实现任意精度级别的精确运动估计。与二次函数拟合（QuadFit）、线性拟合（LinFit）、Sinc拟合（SincFit）、局部质心（LCM）、频域上采样（Upsamp）等算法进行仿真对比，在无噪声污染的情况下，所提算法的平均估计误差为0.0070，运动估计的准确度提高了64%以上；而在有噪声污染的情况下，所提出的算法的平均估计误差为0.0204，运动估计的准确度提高了47%以上。实验结果表明，所提算法不仅能够有效地提高运动估计的精确性，而且具有良好的抗噪性。

关键词: 运动估计, 相位相关, 上采样, 矩阵相乘, 核回归

Abstract: Concerning highly accurate sub-pixel motion vector estimation, an accurate motion estimation algorithm based on upsampled phase correlation with kernel regression refining was proposed. Firstly, an upsampled phase correlation was computed efficiently by means of matrix-multiply discrete Fourier transform, and the initial estimation of motion vector with sub-pixel accuracy was achieved by simply locating its peak. Secondly, a kernel regression function was fit to the upsampled phase correlation values in a neighborhood of initial estimation. Finally, the initial estimation was refined with the location of peak found in the kernel regression fitting function, so as to obtain accurate estimation at arbitrary-precision. In the comparison experiments with some state-of-the-art algorithms such as Quadratic function Fitting (QuadFit), Linear Fitting (LinFit), Sinc Fitting (SincFit), Local Center of Mass (LCM) and Upsampling in the frequency domain (Upsamp), the proposed scheme achieved the average estimation error at 0.0070 in the case of noise-free, and increased the accuracy of motion estimation by more than 64%; while under the noise condition, the average estimation error of the proposed shceme was 0.0204, and the accuracy of motion estimation was improved by more than 47%. Experimental results show that the proposed scheme can not only improve the accuracy of motion estimation significantly, but also achieve good robustness to the influence of noise.

Key words: motion estimation, phase correlation, upsampling, matrix-multiplication, kernel regression

中图分类号:

TP391.41

余应淮, 谢仕义, 梅其祥. 基于核回归修正的上采样相位相关精确运动估计算法[J]. 计算机应用, 2016, 36(8): 2316-2321.

YU Yinghuai, XIE Shiyi, MEI Qixiang. Accurate motion estimation algorithm based on upsampled phase correlation with kernel regression refining[J]. Journal of Computer Applications, 2016, 36(8): 2316-2321.

参考文献

[1] KUGLIN C D.The phase correlation image alignment method[C]//Proceedings of the 1975 IEEE International Conference on Cybernetics and Society.Piscataway,NJ:IEEE,1975:163-165.
[2] ITO K,NIKAIDO A,AOKI T,et al.A dental radiograph recognition system using phase-only correlation for human identification[J].IEICE Transactions on Fundamentals of Electronics,Communications and Computer Sciences,2008,E91-A (1):298-305.
[3] LEPRINCE S,BARBOT S,AYOUB F,et al.Automatic and precise orthorectification,coregistration,and subpixel correlation of satellite images,application to ground deformation measurements[J].IEEE Transactions on Geoscience and Remote Sensing,2007,45(6):1529-1558.
[4] LI X.An improved two-stage image registration algorithm for super-resolution[J].IEEJ Transactions on Electrical and Electronic Engineering,2014,9(4):415-420.
[5] KIM D-H,YOON Y-I,CHOI J-S.An efficient method to build panoramic image mosaics[J].Pattern Recognition Letters,2003,24(14):2421-2429.
[6] ZHANG J,WANG C-S,LIAO W-L.An image mosaics algorithm based on improved phase correlation[C]//ESIAT'09:Proceedings of the 2009 International Conference on Environmental Science and Information Application Technology.Washington,DC:IEEE Computer Society,2009:383-386.
[7] ITO K,NAKAJIMA H,KOBAYASHI K,et al.A fingerprint matching algorithm using phase-only correlation[J].IEICE Transactions on Fundamentals of Electronics,Communications and Computer Sciences,2004,E87-A (3):682-691.
[8] MIYAZAWA K,ITO K,AOKI T,et al.An effective approach for iris recognition using phase-based image matching[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(10):1741-1756.
[9] YOKOYA N,MIYAMURA N,IWASAKI A.Detection and correction of spectral and spatial misregistrations for hyperspectral data using phase correlation method[J].Applied Optics,2010,49(24):4568-4575.
[10] MAY K,KROUGLICOF N.Moving target detection for sense and avoid using regional phase correlation[C]//ICRA 2013:Proceedings of the 2013 IEEE International Conference on Robotics and Automation.Piscataway,NJ:IEEE,2013:4767-4772.
[11] TAKITA K,MUQUIT M,AOKI T,et al.A sub-pixel correspondence search technique for computer vision applications[J].IEICE Transactions on Fundamentals of Electronics,Communications and Computer Sciences,2004,E87-A (8):1913-1923.
[12] OGURI T,IKEHARA M,NGUYEN T.3D cube video coding using phase correlation motion estimation[J].Electronics and Communications in Japan (Part Ⅲ:Fundamental Electronic Science),2006,89(5):32-38.
[13] PAUL M,LIN W,LAU C T,et al.Direct intermode selection for H.264 video coding using phase correlation[J].IEEE Transactions on Image Processing,2011,20(2):461-473.
[14] ALBA A,ARCE-SANTANA E,AGUILAR-PONCE R M,et al.Phase-correlation guided area matching for realtime vision and video encoding[J].Journal of Real-Time Image Processing,2012,9(4):621-633.
[15] PODDER P K,PAUL M,MURSHED M,et al.Fast intermode selection for HEVC video coding using phase correlation[C]//Proceedings of the 2014 International Conference on Digital Image Computing:Techniques and Applications.Piscataway,NJ:IEEE,2015:1-8.
[16] ABDOU I E.Practical approach to the registration of multiple frames of video images[C]//Proceedings of the Visual Communications and Image Processing'99,SPIE 3563.Bellingham,WA:SPIE,1999:371-382.
[17] SHIMIZU M,OKUTOMI M.Sub-pixel estimation error cancellation on area-based matching[J].International Journal of Computer Vision,2005,63(3):207-224.
[18] ARGYRIOU V,VLACHOS T.A study of sub-pixel motion estimation using phase correlation[C]//BMVC 2006:Proceedings of the 200617th British Machine Vision Conference.Edinburgh:British Machine Vision Association,2006:387-396.
[19] HOGE W S.A subspace identification extension to the phase correlation method[J].IEEE Transactions on Medical Imaging,2003,22(2):277-280.
[20] TONG X,YE Z,XU Y,et al.A novel subpixel phase correlation method using singular value decomposition and unified random sample consensus[J].IEEE Transactions on Geoscience and Remote Sensing,2015,53(8):4143-4156.
[21] YU H,CHEN F-S,ZHANG Z-J,et al.A subpixel motion estimation approach based on the phase correlation[C]//OIMT Ⅱ:Proceedings of the 2012 Optoelectronic Imaging and Multimedia Technology Ⅱ,SPIE 8558.Bellingham,WA:SPIE,2012:85580Y.
[22] FOROOSH H,ZERUBIA J B,BERTHOD M.Extension of phase correlation to subpixel registration[J].IEEE Transactions on Image Processing,2002,11(3):188-200.
[23] OUSGUINE S,ESSANOUNI F,ESSANOUNI L,et al.Motion estimation of aliased images using the phase correlation[C]//Proceedings of the 2nd International Conference on Innovative Computing Technology.Washington,DC:IEEE Computer Society,2012:170-173.
[24] GUIZAR-SICAIROS M,THURMAN S T,FIENUP J R.Efficient subpixel image registration algorithms[J].Optics Letters,2008,33(2):156-158.
[25] SOUMMER R,PUEYO L,SIVARAMAKRISHNAN A,et al.Fast computation of Lyot-style coronagraph propagation[J].Optics Express,2007,15(24):15935-15951.
[26] 余应淮,王锦荣.高精度亚像素全局运动估计的上采样梯度互相关算法[J].中国图象图形学报,2012,17(12):1492-1499.(YU Y H,WANG J R.High accuracy sub-pixel global motion estimation based on upsampled gradient cross-correlation algorithm[J].Journal of Image and Graphics,2012,17(12):1492-1499.)
[27] TAKEDA H,FARSIU S,MILANFAR P.Kernel regression for image processing and reconstruction[J].IEEE Transactions on Image Processing,2007,16(2):349-366.
[28] LIU B,LIAO X.Image denoising and magnification via kernel fitting and modified SVD[C]//IAS 2009:Proceedings of the 5th International Conference on Information Assurance and Security.Washington,DC:IEEE Computer Society,2009:521-524.
[29] LIU B-Y,WU W-Y,CHEN X-W.Kernel fitting for image segmentation[C]//ICMLC 2008:Proceedings of the 7th International Conference on Machine Learning and Cybernetics.Washington,DC:IEEE Computer Society,2008,5:2914-2917.
[30] LIU B,ZHANG J.An adaptively trained kernel-based nonlinear representor for handwritten digit classification[J].Journal of Electronics (China),2006,23(3):379-383.

基于核回归修正的上采样相位相关精确运动估计算法

Accurate motion estimation algorithm based on upsampled phase correlation with kernel regression refining

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

[1]	胡高鹏, 陈子鎏, 王晓明, 张开放. 基于深度渐进式反投影注意力网络的图像超分辨率重建[J]. 计算机应用, 2020, 40(7): 2077-2083.
[2]	赵永柱, 黎卫东, 唐斌, 梅峰, 卢文达. 面向期限感知分布式矩阵相乘的高效存储方案[J]. 计算机应用, 2020, 40(2): 311-315.
[3]	樊帆, 高媛, 秦品乐, 王丽芳. 基于并行通道-空间注意力机制的腹部MRI影像多尺度超分辨率重建[J]. 计算机应用, 2020, 40(12): 3624-3630.
[4]	张圆圆, 黄宜军, 王跃飞. 基于纹理信息的室内场景语义标注学习方法[J]. 计算机应用, 2018, 38(12): 3409-3413.
[5]	余应淮, 谢仕义. 滤除椒盐噪声的开关核回归拟合算法[J]. 计算机应用, 2017, 37(10): 2921-2925.
[6]	尹俊, 董利达, 迟天阳. 基于特征点分类策略的移动机器人运动估计[J]. 计算机应用, 2015, 35(2): 590-594.
[7]	庄燕滨桂源肖贤建. 基于运动估计与回溯自适应正交匹配追踪的视频压缩传感[J]. 计算机应用, 2013, 33(09): 2577-2579.
[8]	吴玉莲冯象初. 结合自适应核回归和全变差的乘性噪声去除[J]. 计算机应用, 2013, 33(09): 2592-259.
[9]	景丽张慧娟. 基于相位相关的图像区域复制篡改检测与定位[J]. 计算机应用, 2013, 33(03): 660-662.
[10]	孟勃韩广良. 基于改进的尺度不变特征变换特征点匹配的电子稳像算法[J]. 计算机应用, 2012, 32(10): 2817-2820.
[11]	李世平郑文彬石鑫. H.264运动估计算法UMHexagonS的斐波纳契数列优化[J]. 计算机应用, 2012, 32(09): 2580-2584.
[12]	王凤琴陈晓雷陈燕. 基于时空联合的解码边信息插值算法[J]. 计算机应用, 2012, 32(08): 2324-2327.
[13]	罗桂娥徐运彬. 基于Quaternion-Gabor小波运动估计的深度图获取技术[J]. 计算机应用, 2012, 32(01): 238-240.
[14]	王强李月娥. 基于方向自适应采样搜索的快速运动估计算法[J]. 计算机应用, 2011, 31(10): 2721-2723.
[15]	丁志鸿王刚刘立柱. 基于残差分布的H.264自适应整帧恢复算法[J]. 计算机应用, 2011, 31(06): 1569-1571.