红外视频雷达系统中关键技术的研究
|
摘要
随着图像处理和机器视觉技术的发展,通过双目立体视觉的方法来测量物体的深度距离信息在智能测距系统中的应用越来越广泛。
本文研究的重点是红外视频雷达系统中的两个关键技术的研究。本文首先提出了一个基于数字延时技术的亚像素级匹配方法,然后提出了一个利用流形学习思想来解决摄像机抖动问题的方法。
第一章：介绍立体匹配算法研究的现状,并分析现有算法的一些问题所在。
第二章：论述了立体视觉的一些基础理论,主要包括双目立体视觉的基本原理,摄像机的标定方法以及图像校正方法。
第三章：立体匹配算法原理,以及各种方法。对当前常见的三类立体匹配方法进行介绍。
第四章：详细介绍了基于数字延时技术的亚像素匹配方法。首先,介绍了本匹配方法的整体流程,其次,介绍了数字延时技术的基本原理,最后,阐述了如何将数字延时技术应用于立体匹配。
第五章：主要介绍了利用流形的思想解决摄像机抖动的问题。主要包括抖动问题的分析以及现有方法,流形学习思想以及如何利用流形的思想来达到稳像从而解决运动目标检测中的背景抖动问题。
第六章：总结了本文的研究工作并对未来的后续研究进行了展望。
With the development of technology in image processing and machine vision, measure the depth of objects through the binocular stereo vision method is used in the intelligence system more widely.
Focus of this paper is the two key technologies of infrared video radar system. First, this paper presents a digital delay technology based sub-pixel matching method. Second, presents a method that uses manifold learning to solve the jitter of camera.
Chapter 1:Introduction to the status of stereo matching algorithm, and analyze some of the problems of existing algorithms.
Chapter 2:The paper discusses some of the basic theory of stereo vision, including the basic principle of binocular stereo vision, camera calibration and image correction method.
Chapter 3:Principles of stereo matching algorithms and a variety of methods. Three categories of current common stereo matching are introduced.
Chapter 4:Description of technology based on digital delay sub-pixel matching. First of all, the overall matching process is introduced, and second, the basic principles of digital delay technology are described, finally, how to use the digital delay technology in stereo match is described.
Chapter 5:The method for solving the problem of the camera shaking is described. In this chapter, we discuss the problems of jitter and the existing methods, how to solve the problem of the background jitter in moving target detection using the idea of manifold.
Chapter 6:Summary of the research work of this paper and the future prospect of the follow-up study will be carried out.
本文研究的重点是红外视频雷达系统中的两个关键技术的研究。本文首先提出了一个基于数字延时技术的亚像素级匹配方法,然后提出了一个利用流形学习思想来解决摄像机抖动问题的方法。
第一章：介绍立体匹配算法研究的现状,并分析现有算法的一些问题所在。
第二章：论述了立体视觉的一些基础理论,主要包括双目立体视觉的基本原理,摄像机的标定方法以及图像校正方法。
第三章：立体匹配算法原理,以及各种方法。对当前常见的三类立体匹配方法进行介绍。
第四章：详细介绍了基于数字延时技术的亚像素匹配方法。首先,介绍了本匹配方法的整体流程,其次,介绍了数字延时技术的基本原理,最后,阐述了如何将数字延时技术应用于立体匹配。
第五章：主要介绍了利用流形的思想解决摄像机抖动的问题。主要包括抖动问题的分析以及现有方法,流形学习思想以及如何利用流形的思想来达到稳像从而解决运动目标检测中的背景抖动问题。
第六章：总结了本文的研究工作并对未来的后续研究进行了展望。
With the development of technology in image processing and machine vision, measure the depth of objects through the binocular stereo vision method is used in the intelligence system more widely.
Focus of this paper is the two key technologies of infrared video radar system. First, this paper presents a digital delay technology based sub-pixel matching method. Second, presents a method that uses manifold learning to solve the jitter of camera.
Chapter 1:Introduction to the status of stereo matching algorithm, and analyze some of the problems of existing algorithms.
Chapter 2:The paper discusses some of the basic theory of stereo vision, including the basic principle of binocular stereo vision, camera calibration and image correction method.
Chapter 3:Principles of stereo matching algorithms and a variety of methods. Three categories of current common stereo matching are introduced.
Chapter 4:Description of technology based on digital delay sub-pixel matching. First of all, the overall matching process is introduced, and second, the basic principles of digital delay technology are described, finally, how to use the digital delay technology in stereo match is described.
Chapter 5:The method for solving the problem of the camera shaking is described. In this chapter, we discuss the problems of jitter and the existing methods, how to solve the problem of the background jitter in moving target detection using the idea of manifold.
Chapter 6:Summary of the research work of this paper and the future prospect of the follow-up study will be carried out.
引文
[1]张广军：机器视觉；北京,科学出版社,2005
[2]Marr D著.姚国正译：视觉计算理论；北京,科学出版社,1988
[3]游素亚,徐光祜：立体视觉研究的现状与进展；中国图象图形学报,1997,2(1)：17-24
[4]Mayhew.etal.Psychophysical and computational study towards a theory of human stereopsis[J] Artificial Intelligence,1981,17:349-385.
[5]徐奕,周军,周源华：立体视觉匹配技术[J]；计算机工程与应用,2003,15：1-5
[6]Scharstein D, Szeliski R. A taxonomy and eValuation of dense two-frame stereo correspondence algorithm [J].International Journal of Computer Vision,2002,47(1/2/3):7-42.
[7]隋婧,金伟其：双目立体视觉技术的实现及其进展；电子技术应用网,2008年8月
[8]章毓晋.图像工程(下册)[M].北京：清华大学出版社,2007：103-105.
[9]马颂德,张正友.计算机视觉—计算理论与算法基础.北京：科学出版社,1998：52-71
[10]R.Tsai, An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision. Proc of Conference of Computer Vision and Pattern Recognition.1986:364-374.
[11]R.I.Hartley.Theory and practice of projective rectification.Technical Report2538,INRIA,April 1995.
[12]O.Faugeras.Three-Dimensional computer vision:a geometric viewpoint.MIT Press,1993.
[13]S.B.Kang,J.A.Webb,Z.C.L,and T.Kanade.An active multibaseline stereo system with real-time image acquisition.Technical Report CMU-CS-94-167,School of Computer Science,Carnegie Mellon University,1994
[14]宰小涛：基于SIFT特征描述子的立体匹配算法研究；上海交通大学硕士学位论文,2007
[15]刘欣.ALV立体视觉导航技术研究：[硕士学位论文],国防科技大学,2005
[16]郑飞.基于立体视觉的三维场景重建与绘制：[硕士学位论文],合肥工业大学,2003
[17]Barnard S,etc.Disparity analysis of images.IEEE Trans.on PAMI,1980,2:33-340
[18]RAYMOND V E,CLIFTTON M S. Unconstrained stereoscopic matching of lines [J].Vision Research,200,40:151-162
[19]KANADE T,OKUTOMI M.A stereo matching algorithm with an adaptive window [J].Theory and Experiment,1994,16(9):920-931
[20]李德广,李科杰,高丽丽：基于多尺度多方向相伴匹配的立体视觉方法[J].仪器仪表学报,2004,25(4)：600-602.
LI D G, LI K J, GAO L L:Stereo vision using multiresolution and multiorientation phase matching[J].Chinese Journal of Scientific Instrument,2004,25(4):600-602.
[21]Wang B,Shen C L:Dense disparity map estimation using genetic algorithms.Trans of Nanjing University of Aeronautics & Astronautics,2003,20(2):184-191.
[22]徐奕,周军,周源华.立体视觉匹配技术.计算机工程与应用,2003,15：1-5
[23]任重,邵军力,双目立体视觉的匹配方法研究,军事通信技术,2001,22(2)：45—48
[24]Kim H, Sohn K.Hierarchical disparity estimation with energy-based regularization[C].Proc of Int Conf on Image Processing.Barcelona,2003:373-376.
[25]Prince J D, Eagle R A.Weighted directional energy model of human stereo correspondence[J].Vision Research,2000,40(9):1143-1155.
[26]Kuglin C D,Hines D C.The phase correlation image alignment method[C].Proc.of IEEE Int. Conf.on Cybernetics and Society,1975:163-165
[27]Fleet D J,Jepson A D,Jenkin R M.Phase-based disparity measurement.CVGIP:Image understanding,1991,53(2):198-210
[28]Zhou J,Xu Y,Yu W R.Phase matching with multi-resolution wavelet transform. Proceeding. SPIE,Bellingham,2002,61 (10):82-91.
[29]于慧敏等：信号与系统(第二版)；北京：化学工业出版社,2007：192-197
[30]王肠,冯驰：基于块匹配的电子图像稳定算法；咸阳师范学院学报,2006年8月
[31]赵红颖,熊经武：获取动态图像位移矢量的灰度投影算法的应用[J].光电工程,2001,6(28)：51-54
[32]孙辉,赵红颖,熊经武,金宏：基于光流模型的图像运动估计方法[J].光学精密工程,2002,443-447
[33]Belkin M, Niyogi P.Laplacian eigenmaps for dimensionality reduction and data representation J].Neural Computation,2003,15(6):1373-1396.
[34]罗四维,赵连伟：基于谱图理论的流形学习算法[J].计算机研究与发展,2006,43(7)：1173-1179.
[35]Tenenbaum J B,de Silva V,Langford J C .A global geometric frame-work for nonlinear dimensionality reduction [J].Science,2000,290(5500):2319-2323
[36]Roweis S,Sal L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290(5500):2323-2326
[37]He X,Niyogi P.Locality preserving projections[C]//Advances in Neural Information Processing Systems.Cambrige:MIT Press,2003:291-299
[38]Yan S,Xu D,zhang B,et al.Graph embedding and extensions:a general framework for dimensionality reduction[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2007,29(1):40-51
[39]Lowe D G. Distinctive image features from scale-invariant interest points [J].International Journal of Computer Vision,2004,60(2):91-110
[2]Marr D著.姚国正译：视觉计算理论；北京,科学出版社,1988
[3]游素亚,徐光祜：立体视觉研究的现状与进展；中国图象图形学报,1997,2(1)：17-24
[4]Mayhew.etal.Psychophysical and computational study towards a theory of human stereopsis[J] Artificial Intelligence,1981,17:349-385.
[5]徐奕,周军,周源华：立体视觉匹配技术[J]；计算机工程与应用,2003,15：1-5
[6]Scharstein D, Szeliski R. A taxonomy and eValuation of dense two-frame stereo correspondence algorithm [J].International Journal of Computer Vision,2002,47(1/2/3):7-42.
[7]隋婧,金伟其：双目立体视觉技术的实现及其进展；电子技术应用网,2008年8月
[8]章毓晋.图像工程(下册)[M].北京：清华大学出版社,2007：103-105.
[9]马颂德,张正友.计算机视觉—计算理论与算法基础.北京：科学出版社,1998：52-71
[10]R.Tsai, An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision. Proc of Conference of Computer Vision and Pattern Recognition.1986:364-374.
[11]R.I.Hartley.Theory and practice of projective rectification.Technical Report2538,INRIA,April 1995.
[12]O.Faugeras.Three-Dimensional computer vision:a geometric viewpoint.MIT Press,1993.
[13]S.B.Kang,J.A.Webb,Z.C.L,and T.Kanade.An active multibaseline stereo system with real-time image acquisition.Technical Report CMU-CS-94-167,School of Computer Science,Carnegie Mellon University,1994
[14]宰小涛：基于SIFT特征描述子的立体匹配算法研究；上海交通大学硕士学位论文,2007
[15]刘欣.ALV立体视觉导航技术研究：[硕士学位论文],国防科技大学,2005
[16]郑飞.基于立体视觉的三维场景重建与绘制：[硕士学位论文],合肥工业大学,2003
[17]Barnard S,etc.Disparity analysis of images.IEEE Trans.on PAMI,1980,2:33-340
[18]RAYMOND V E,CLIFTTON M S. Unconstrained stereoscopic matching of lines [J].Vision Research,200,40:151-162
[19]KANADE T,OKUTOMI M.A stereo matching algorithm with an adaptive window [J].Theory and Experiment,1994,16(9):920-931
[20]李德广,李科杰,高丽丽：基于多尺度多方向相伴匹配的立体视觉方法[J].仪器仪表学报,2004,25(4)：600-602.
LI D G, LI K J, GAO L L:Stereo vision using multiresolution and multiorientation phase matching[J].Chinese Journal of Scientific Instrument,2004,25(4):600-602.
[21]Wang B,Shen C L:Dense disparity map estimation using genetic algorithms.Trans of Nanjing University of Aeronautics & Astronautics,2003,20(2):184-191.
[22]徐奕,周军,周源华.立体视觉匹配技术.计算机工程与应用,2003,15：1-5
[23]任重,邵军力,双目立体视觉的匹配方法研究,军事通信技术,2001,22(2)：45—48
[24]Kim H, Sohn K.Hierarchical disparity estimation with energy-based regularization[C].Proc of Int Conf on Image Processing.Barcelona,2003:373-376.
[25]Prince J D, Eagle R A.Weighted directional energy model of human stereo correspondence[J].Vision Research,2000,40(9):1143-1155.
[26]Kuglin C D,Hines D C.The phase correlation image alignment method[C].Proc.of IEEE Int. Conf.on Cybernetics and Society,1975:163-165
[27]Fleet D J,Jepson A D,Jenkin R M.Phase-based disparity measurement.CVGIP:Image understanding,1991,53(2):198-210
[28]Zhou J,Xu Y,Yu W R.Phase matching with multi-resolution wavelet transform. Proceeding. SPIE,Bellingham,2002,61 (10):82-91.
[29]于慧敏等：信号与系统(第二版)；北京：化学工业出版社,2007：192-197
[30]王肠,冯驰：基于块匹配的电子图像稳定算法；咸阳师范学院学报,2006年8月
[31]赵红颖,熊经武：获取动态图像位移矢量的灰度投影算法的应用[J].光电工程,2001,6(28)：51-54
[32]孙辉,赵红颖,熊经武,金宏：基于光流模型的图像运动估计方法[J].光学精密工程,2002,443-447
[33]Belkin M, Niyogi P.Laplacian eigenmaps for dimensionality reduction and data representation J].Neural Computation,2003,15(6):1373-1396.
[34]罗四维,赵连伟：基于谱图理论的流形学习算法[J].计算机研究与发展,2006,43(7)：1173-1179.
[35]Tenenbaum J B,de Silva V,Langford J C .A global geometric frame-work for nonlinear dimensionality reduction [J].Science,2000,290(5500):2319-2323
[36]Roweis S,Sal L K.Nonlinear dimensionality reduction by locally linear embedding[J].Science,2000,290(5500):2323-2326
[37]He X,Niyogi P.Locality preserving projections[C]//Advances in Neural Information Processing Systems.Cambrige:MIT Press,2003:291-299
[38]Yan S,Xu D,zhang B,et al.Graph embedding and extensions:a general framework for dimensionality reduction[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2007,29(1):40-51
[39]Lowe D G. Distinctive image features from scale-invariant interest points [J].International Journal of Computer Vision,2004,60(2):91-110