三维粒子跟踪测速系统中的三维重构技术研究
|
摘要
基于视觉的三维粒子图像测速技术(Three-dimensional Particle Image Velocimetry,3DPIV)和三维粒子跟踪测速技术(Three-dimensional Particle Tracking Velocimetry,3D PTV)较传统流场测量技术在流场测量空间分辨率、流场可测量体积等多个方面有了极大的提升,其已成为流场测量技术领域中的研究重点和热点。3D PIV可获得较高的流场测量空间分辨率,但粒子间的重叠问题限制了其可测量流场体积。3D PTV可对更大体积的流场进行测量,但其流场测量空间分辨率不高。在对较大体积的流场进行测量时,3D PTV较3D PIV具有更高的优越性。为使3D PTV具有更高的流场测量空间分辨率,须提高其示踪粒子浓度。但示踪粒子浓度升高将致使3D PTV在三维粒子重构和三维粒子运动轨迹重构等方面的难度急剧增加。鉴于此,本文对3D PTV中的三维重构技术进行了探索和研究。
首先,本文将离焦成像技术和离焦深度测量技术引入到3D PTV中,以粒子离焦半径作为粒子的另一特征,并利用离焦深度测量技术对粒子深度进行估计。通过对粒子离焦成像原理进行分析,基于离焦深度测量技术提出了一种粒子深度测量及标定方法,其利用粒子离焦半径可计算出粒子深度及其不确定度。为对离焦粒子中心及半径进行准确检测,本文分别基于圆的霍夫变换和距离变换提出了两种离焦粒子中心及半径检测方法,并对两种方法的粒子中心及半径检测性能进行了实验分析。实验结果表明,本文提出的两种方法均可对离焦粒子图像中的无重叠粒子和重叠粒子进行准确地检测,但基于距离变换的粒子中心及半径检测方法具有更高的运算效率。
其次,利用已获得的粒子深度测量及标定方法,本文分别基于单目视觉技术和双目视觉技术提出了两种三维粒子重构方法。由于光线折射的存在,粒子成像系统观测到的粒子为真实粒子的虚像,粒子深度测量方法计算出的粒子深度也为粒子虚像深度。为获取真实粒子深度,本文对真实粒子深度同其虚像深度间的几何光学关系进行了分析,并建立了数学关系模型。由于粒子离焦深度测量及标定方法存在着一定不确定度,所以基于单目视觉的三维粒子重构方法的三维重构精度不高。为进一步提高三维粒子重构精度,在基于单目视觉和离焦深度测量的三维粒子重构方法的基础上,本文提出了一种基于双目视觉的三维粒子重构方法。为提高粒子立体匹配正确率,将粒子深度及其不确定度引入到粒子立体匹配分析中,并提出了极线段约束和粒子离焦半径约束两种粒子立体匹配约束条件。通过应用新的粒子立体匹配约束条件,有效地提高了粒子立体匹配正确率。为准确地重构出粒子三维空间位置,在考虑光折射影响的前提下,建立了基于双目视觉的三维粒子重构几何模型。
再次,为提高3D PTV的可靠性,并降低3D PTV算法实现难度,通过借鉴3D PIV中的流场描述方法,本文采用三维粒子运动矢量替代粒子三维运动轨迹对流场进行描述。为对三维粒子运动矢量进行准确重构,本文在基于双目视觉的三维粒子重构方法的基础上,提出了一种基于双目视觉的三维粒子运动矢量重构方法。针对粒子运动矢量立体匹配问题,提出了一种粒子运动矢量立体匹配约束。通过应用粒子运动矢量立体匹配约束,粒子运动矢量立体匹配正确率得到有效的提高。同时,为解决流场中可能存在的小区域内粒子浓度过高的问题,将基于深度剖层的3D PIV方法引入到了三维粒子运动矢量重构中,并针对其互相关运算量较大的问题改进了基于FFT的互相关算法。
最后,为对本文所提出的理论和方法进行综合性分析和验证,分别构建了仿真流场测量实验系统和真实流场测量实验系统,并进行了仿真流场测量实验和真实流场测量实验。仿真实验结果表明本文方法较传统3D PTV方法在三维粒子和三维粒子运动矢量的重构比例和重构精度方面均有了一定的提高。在真实流场测量实验中,分别构建了基于单目视觉和双目视觉的三维粒子跟踪测速系统,并在严格和非严格实验环境下进行了流场测量实验。真实流场测量实验结果表明本文所提方法具有较强的可行性和可用性。
Compared with the invasive flow measurement methods, the performances of the three-dimensional particle image velocimetry (3D PIV) and the three-dimensional particle tracking velocimetry (3D PTV), such as the spatial resolution of flow field and the observable volume, are significantly improved. Hereby, they have been widely researched and more attention was paid to the development of new3D PIV and3D PTV methods. The spatial resolution of flow field measured by3D PIV is higher for the high particle density compared to3D PTV, but the observable volume is limited because of the particle overlapping issue. Due to the observable volume limitation in3D PIV,3D PTV is usually employed as the major method to analyze the large volume flow field with the low particle density. To improve the spatial resolution of flow field measured by3D PTV, the particle density must be increased. As a result, the difficulties of the three-dimensional particle reconstruction and the three-dimensional particle trajectory reconstruction get harder to be solved. For these reasons, much work was carried out to try to solve the issues of3D PTV in this research.
Firstly, the defocus imaging and the depth from defocus techniques are introduced into the3D PTV, and the blur circle radius is used as the feature of particle. According to the defocus imaging and the depth from defocus techniques, a particle depth fitting and measuring method was proposed to calculate the particle depth and its uncertainty by using the particle blur circle radius. To determine the particle center and blur circle radius, two particle determination methods based on Circular Hough Transform and Distance Transform were proposed separately, and the particle determination experiments were conducted. The experimental results show that the particle center and blur circle radius were determined accurately by both methods. In addition, the Distance Transform based particle determination method is more efficient.
Secondly, according to the proposed particle depth fitting and measuring method, two three-dimensional particle reconstruction methods based on the singular vision technology and the binocular vision technology were proposed separately. Due to the light refraction occurred on the interface between the different mediums, the particle captured by the singular particle imaging system is the virtual image of the real particle. The particle depth computed by the proposed particle depth fitting and measuring method is the depth of virtual particle. To recover the depth of real particle, the geometric relationship between the real particle and the virtual particle was deduced. Due to the uncertainty of estimated particle depth, the accuracy of3D particle reconstruction of the singular vision based3D particle reconstruction is low. To improve the accuracy of3D particle reconstruction, a new binocular vision based3D particle reconstruction method was proposed. The particle depth and the particle depth uncertainty computed by using the singular vision based3D particle reconstruction method are inducted into the binocular particle pairing analysis, and two binocular particle pairing constraints, which are the epipolar line segment constraint and the defocused particle radius constraint, were presented to reduce the number of pairing particle candidates and improve the rate of correct particle pairing. And, a binocular vision based three-dimensional particle reconstruction method was presented with considering the occurred light refraction.
Thirdly, due to the difficulty of three-dimensional particle trajectory reconstruction in the traditional3D PTV, the three-dimensional particle vector is employed instead of the three-dimensional particle trajectory to measure the flow filed by referring the3D PIV concepts. And, a new3D particle vector reconstruction method was proposed to solve the difficulties in the binocular particle vector matching. By using the proposed binocular particle vector pairing method, most of the falsely paired particles are removed and the3D particle vectors are reconstructed correctly. Moreover, the depth tomography method is introduced into the3D particle vector reconstruction for solving the issue induced by the high particle density in a small region of the observable volume. Furthermore, the computational efficiency of the FFT based cross-correlation algorithm is improved by using the decimation in frequency theory.
Finally, for validating the presented method, the simulation flow field measurement system and the actual flow field measurement system were constructed separately. And, many experiments were performed with the constructed simulation and actual flow field measurement systems. The simulation experiments of flow field measurement were conducted with the proposed method and the traditional3D PTV method. The results of the simulation experiments are shown that the number and the accuracy of the reconstructed3D particle and particle vector by using the proposed method were improved compared with the traditional3D PTV method. And, the actual flow filed measurement experiments were conducted with the constructed singular vision and binocular vision based3D PTV systems under the constrained and unconstrained experimental conditions. The results of the actual experiments are shown that the proposed methods are applicable to measure the flow field.
首先,本文将离焦成像技术和离焦深度测量技术引入到3D PTV中,以粒子离焦半径作为粒子的另一特征,并利用离焦深度测量技术对粒子深度进行估计。通过对粒子离焦成像原理进行分析,基于离焦深度测量技术提出了一种粒子深度测量及标定方法,其利用粒子离焦半径可计算出粒子深度及其不确定度。为对离焦粒子中心及半径进行准确检测,本文分别基于圆的霍夫变换和距离变换提出了两种离焦粒子中心及半径检测方法,并对两种方法的粒子中心及半径检测性能进行了实验分析。实验结果表明,本文提出的两种方法均可对离焦粒子图像中的无重叠粒子和重叠粒子进行准确地检测,但基于距离变换的粒子中心及半径检测方法具有更高的运算效率。
其次,利用已获得的粒子深度测量及标定方法,本文分别基于单目视觉技术和双目视觉技术提出了两种三维粒子重构方法。由于光线折射的存在,粒子成像系统观测到的粒子为真实粒子的虚像,粒子深度测量方法计算出的粒子深度也为粒子虚像深度。为获取真实粒子深度,本文对真实粒子深度同其虚像深度间的几何光学关系进行了分析,并建立了数学关系模型。由于粒子离焦深度测量及标定方法存在着一定不确定度,所以基于单目视觉的三维粒子重构方法的三维重构精度不高。为进一步提高三维粒子重构精度,在基于单目视觉和离焦深度测量的三维粒子重构方法的基础上,本文提出了一种基于双目视觉的三维粒子重构方法。为提高粒子立体匹配正确率,将粒子深度及其不确定度引入到粒子立体匹配分析中,并提出了极线段约束和粒子离焦半径约束两种粒子立体匹配约束条件。通过应用新的粒子立体匹配约束条件,有效地提高了粒子立体匹配正确率。为准确地重构出粒子三维空间位置,在考虑光折射影响的前提下,建立了基于双目视觉的三维粒子重构几何模型。
再次,为提高3D PTV的可靠性,并降低3D PTV算法实现难度,通过借鉴3D PIV中的流场描述方法,本文采用三维粒子运动矢量替代粒子三维运动轨迹对流场进行描述。为对三维粒子运动矢量进行准确重构,本文在基于双目视觉的三维粒子重构方法的基础上,提出了一种基于双目视觉的三维粒子运动矢量重构方法。针对粒子运动矢量立体匹配问题,提出了一种粒子运动矢量立体匹配约束。通过应用粒子运动矢量立体匹配约束,粒子运动矢量立体匹配正确率得到有效的提高。同时,为解决流场中可能存在的小区域内粒子浓度过高的问题,将基于深度剖层的3D PIV方法引入到了三维粒子运动矢量重构中,并针对其互相关运算量较大的问题改进了基于FFT的互相关算法。
最后,为对本文所提出的理论和方法进行综合性分析和验证,分别构建了仿真流场测量实验系统和真实流场测量实验系统,并进行了仿真流场测量实验和真实流场测量实验。仿真实验结果表明本文方法较传统3D PTV方法在三维粒子和三维粒子运动矢量的重构比例和重构精度方面均有了一定的提高。在真实流场测量实验中,分别构建了基于单目视觉和双目视觉的三维粒子跟踪测速系统,并在严格和非严格实验环境下进行了流场测量实验。真实流场测量实验结果表明本文所提方法具有较强的可行性和可用性。
Compared with the invasive flow measurement methods, the performances of the three-dimensional particle image velocimetry (3D PIV) and the three-dimensional particle tracking velocimetry (3D PTV), such as the spatial resolution of flow field and the observable volume, are significantly improved. Hereby, they have been widely researched and more attention was paid to the development of new3D PIV and3D PTV methods. The spatial resolution of flow field measured by3D PIV is higher for the high particle density compared to3D PTV, but the observable volume is limited because of the particle overlapping issue. Due to the observable volume limitation in3D PIV,3D PTV is usually employed as the major method to analyze the large volume flow field with the low particle density. To improve the spatial resolution of flow field measured by3D PTV, the particle density must be increased. As a result, the difficulties of the three-dimensional particle reconstruction and the three-dimensional particle trajectory reconstruction get harder to be solved. For these reasons, much work was carried out to try to solve the issues of3D PTV in this research.
Firstly, the defocus imaging and the depth from defocus techniques are introduced into the3D PTV, and the blur circle radius is used as the feature of particle. According to the defocus imaging and the depth from defocus techniques, a particle depth fitting and measuring method was proposed to calculate the particle depth and its uncertainty by using the particle blur circle radius. To determine the particle center and blur circle radius, two particle determination methods based on Circular Hough Transform and Distance Transform were proposed separately, and the particle determination experiments were conducted. The experimental results show that the particle center and blur circle radius were determined accurately by both methods. In addition, the Distance Transform based particle determination method is more efficient.
Secondly, according to the proposed particle depth fitting and measuring method, two three-dimensional particle reconstruction methods based on the singular vision technology and the binocular vision technology were proposed separately. Due to the light refraction occurred on the interface between the different mediums, the particle captured by the singular particle imaging system is the virtual image of the real particle. The particle depth computed by the proposed particle depth fitting and measuring method is the depth of virtual particle. To recover the depth of real particle, the geometric relationship between the real particle and the virtual particle was deduced. Due to the uncertainty of estimated particle depth, the accuracy of3D particle reconstruction of the singular vision based3D particle reconstruction is low. To improve the accuracy of3D particle reconstruction, a new binocular vision based3D particle reconstruction method was proposed. The particle depth and the particle depth uncertainty computed by using the singular vision based3D particle reconstruction method are inducted into the binocular particle pairing analysis, and two binocular particle pairing constraints, which are the epipolar line segment constraint and the defocused particle radius constraint, were presented to reduce the number of pairing particle candidates and improve the rate of correct particle pairing. And, a binocular vision based three-dimensional particle reconstruction method was presented with considering the occurred light refraction.
Thirdly, due to the difficulty of three-dimensional particle trajectory reconstruction in the traditional3D PTV, the three-dimensional particle vector is employed instead of the three-dimensional particle trajectory to measure the flow filed by referring the3D PIV concepts. And, a new3D particle vector reconstruction method was proposed to solve the difficulties in the binocular particle vector matching. By using the proposed binocular particle vector pairing method, most of the falsely paired particles are removed and the3D particle vectors are reconstructed correctly. Moreover, the depth tomography method is introduced into the3D particle vector reconstruction for solving the issue induced by the high particle density in a small region of the observable volume. Furthermore, the computational efficiency of the FFT based cross-correlation algorithm is improved by using the decimation in frequency theory.
Finally, for validating the presented method, the simulation flow field measurement system and the actual flow field measurement system were constructed separately. And, many experiments were performed with the constructed simulation and actual flow field measurement systems. The simulation experiments of flow field measurement were conducted with the proposed method and the traditional3D PTV method. The results of the simulation experiments are shown that the number and the accuracy of the reconstructed3D particle and particle vector by using the proposed method were improved compared with the traditional3D PTV method. And, the actual flow filed measurement experiments were conducted with the constructed singular vision and binocular vision based3D PTV systems under the constrained and unconstrained experimental conditions. The results of the actual experiments are shown that the proposed methods are applicable to measure the flow field.
引文
[1]Sheng J, Meng H, Fox R O. Validation of CFD simulations of a stirred tank using Particle Image Velocimetry data[J]. The Canadian Journal of Chemical Engineering.1998,76(3):611-625.
[2]Raffel M, Willert C E, Wereley S T, et al. Particle image velocimetry: a practical guide (Experimental Fluid Mechanics)[M]. New York: Springer,2007:8-59.
[3]Maas H G, Gruen A, Papantoniou D. Particle tracking velocimetry in three-dimensional flows[J]. Experiments in Fluids.1993,15(2):133-146.
[4]Arroyo M, Hinsch K. Recent developments of PIV towards 3D measurements Particle Image Velocimetry[M]. Berlin: Springer,2008:127-154.
[5]Haigermoser C, Scarano F, Onorato M. Investigation of the flow in a circular cavity using stereo and tomographic particle image velocimetry [J]. Experiments in Fluids.2009,46(3):517-526.
[6]Coudert S, Foucaut J, Kostas J, et al. Double large field stereoscopic PIV in a high reynolds number turbulent boundary layer[J]. Experiments in Fluids.2011,50(1):1-12.
[7]Berg E J, Robinson R J. Stereoscopic particle image velocimetry analysis of healthy and emphysemic alveolar sac models[J]. Journal of Biomechanical Engineering.2011, 133(6):061004-061008.
[8]Spence C, Buchmann N, Jermy M, et al. Stereoscopic PIV measurements of flow in the nasal cavity with high flow therapy[J]. Experiments in Fluids.2011,50(4):1005-1017.
[9]Willneff J. Spatio-temporal matching algorithm for 3D particle tracking velocimetry[D]. Zurich: Swiss Federal Institute of Technology Zuric,2002.
[10]Kieft R K, Schreel K S, van der Plas GvdP, et al. The application of a 3D PTV algorithm to a mixed convection flow[J]. Experiments in Fluids.2002,33(4):603-611.
[11]Maas H G, Putze T, Westfeld P. Recent developments in 3D-PTV and Tomo-PIV[J]. Imaging Measurement Methods for Flow Analysis.2009,106(1):53-62.
[12]Peterson S D, Chuang H S, Wereley S T. Three-dimensional particle tracking using micro-particle image velocimetry hardware[J]. Measurement Science and Technology.2008, 19(11):1-8.
[13]Peterson K, Regaard B, Heinemann S, et al. Single-camera, three-dimensional particle tracking velocimetry[J]. Optics Express.2012,20(8):9031-9037.
[14]Tarlet D, Bendicks C, Roloff C, et al. Gas flow measurements by 3D particle Tracking velocimetry using coloured tracer particles[J]. Flow, Turbulence and Combustion,2012, 88(3):343-365.
[15]Patel R S, Garimella S V. Development of a particle tracking-based measurement technique to map three-dimensional interfaces between transparent, immiscible fluids[C].13th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems. San Diego, USA,2012:1432-1437.
[16]Ushijima S, Tanaka N. Three-dimensional particle tracking velocimetry with laser-light sheet scanning[J]. Journal of Fluids Engineering.1996,118(2):352-357.
[17]Buchmann N, Atkinson C, Jeremy M, et al. Tomographic particle image velocimetry investigation of the flow in a modeled human carotid artery bifurcation[J]. Experiments in Fluids. 2011,50(4):1131-1151.
[18]Wegfrass A, Lobutova E, Resagk C. Application of 3D particle tracking velocimetry in convection[J]. Springer Proceedings in Physics.2012,141:297-300.
[19]Malek M, Lebrun D, Allano D. Digital in-line holography system for 3D-3C particle tracking velocimetry[J]. Particle Image Velocimetry.2008,112(1):155-170.
[20]Dubsky S, Jamison R, Higgins S, et al. Computed tomographic X-ray velocimetry for simultaneous 3D measurement of velocity and geometry in opaque vessels[J]. Experiments in Fluids.2012,52(3):543-554.
[21]Seifi M, Fournier C, Denis L, et al. Three-dimensional reconstruction of particle holograms:a fast and accurate multiscale approach[J]. Journal of the Optical Society of America.2012, 29(9):1808-1817.
[22]Okamoto K, Schmidl W D, Hassan Y A, Spring model tracking algorithm for three-dimensional particle image velocimetry[J]. ASME Fluids Engineering Division.1995,209:91-97.
[23]Okamoto K, Three-dimensional particle cluster tracking algorithm using affine transformation[J]. Journal of Visualization.1998,1 (2):153-160.
[24]Ishikawa M, Murai Y, Wada A, et al. A novel algorithm for particle tracking velocimetry using the velocity gradient tensor tensor[J]. Experiments in Fluids.2000,29(6):519-531
[25]Ohyama R, Takagi T, Tsukiji T, et al. Particle tracking technique and velocity measurement of visualized flow fields by means of genetic algorithm[J]. Journal of the Visualization Society.1993, 13(S1):35-38.
[26]Kimura I, Hattori A, Ueda M. Particle airing using genetic algorithms for PIV[J]. Journal of Visualization.2000,2(3-4):223-228.
[27]Grant I, Pan X. An investigation of the performance of multi-layer neural networks applied to the analysis of PI V images [J]. Experiments in Fluids.1995,19(3):159-166.
[28]Knaak M. A hopfield neural network for flow field computation based on particle image velocimetry/particle tracking velocimetry image sequences[C], IEEE International Conference on Neural Networks. Houston, USA,1997:48-52.
[29]Ohmi K, Sapkota A. Particle tracking velocimetry using cellular neural networks[C]. International Joint Conference on Neural Networks. Vancouver, Canada,2006:3963-3969.
[30]Labonte G. A new neural network for particle tracking velocimetry[J]. Experiments in Fluids. 1999,26(4):340-346
[31]Wernet M P. Fuzzy logic particle tracking velocimetry[C]. Proceeding of SPIE on Optical Diagnostics in Fluid and Thermal Flow. San Diego, USA,2006:198:200.
[32]Takagi T. Study on particle tracking velocimetry using ant colony optimization [J], Journal of Visualization Society.2007,27(S2):89-90.
[33]Ohmi K, Sapkota A, Panday S P. Applicability of ant colony optimization in particle tracking velocimetry[C].14th International Symposium on Applications of Laser Techniques to Fluid Mechanics. Lisbon, Portugal,2008:1-12.
[34]田晓东,李玉梁,陈嘉范,等.一种新改进的DPIV方法[J].水科学进展.2000,11(1)：59-63.
[35]陈波,李万平.粒子图像测速互相关中亚像素位移定位的改进方法[J].计算机辅助设计与图形学学报.2011,23(11)：1896-1901.
[36]王希麟,张大力,常辙,等.两相流场粒子成像测速技术(PTV-PIV)初探[J].力学学报.1998,30(1)：121-125.
[37]康琦,申功忻.全场测速技术进展[J].力学进展.1997,27(1)：106-121.
[38]许联锋.水气两相流动的数字图像测量方法及应用研究[D],西安：西安理工大学,2004.
[39]白玉川.复杂表面流场中粒子跟踪测速技术的研究[C],第二十届全国水动力学研讨会,太原：《水动力学研究与进展》编委会,2007.
[40]宋连壮.基于图像处理的气固稀相流场测量及气泡行为[D],石家庄：东北电力大学,2011.
[41]周云龙,宋连壮,周红娟.基于特征相似度匹配PTV法的稀相气固两相流流场的测量[J].化工自动化及仪表.2010,37(10)：58-61.
[42]金家莉.二维数字滤波在物理模型流场测量系统中的应用[J].水道港口.2007,28(6)：448-452.
[43]禹明忠.PTV技术和颗粒三维运动规律的研究[D],北京：清华大学,2002.
[44]蔡毅,由长福,祁海鹰,等.模糊逻辑方法用于气固两相流动PTV测量中的颗粒识别过程[J].液体力学实验与测量.2002,16(2)：78-83.
[45]靳斌,杨冠玲,何振江,等.一种利用示踪粒子群体运动特征的PTV方法[J].光学技术.2000,26(1)：16-18.
[46]杜海,李木国.三维粒子测速系统中遗传匹配法的设计[J].系统仿真学报.2008,20(5)：1251-1258.
[47]靳斌,何振江.一种单相机流场速度场三维测量方法[J].光学技术.2000,26(5)：4450450.
[48]罗锐,孙艳飞,杨献勇.应用于微细流场的全息荧光粒子图像三维测速系统[J].工程热物理学报.27(3)：493-495.
[49]吕且妮,葛宝臻,高岩等.乙醇喷雾场粒子尺寸和速度的数字全息测量[J].光子学报.39(2)：266-270.
[50]Hassan Y, Blanchat T. Full-field bubbly flow velocity measurements by digital image pulsed laser velocimetry[J]. Experiments in Fluids.1991,11(5):293-301.
[51]Ortiz-Villafuerte J, Schmidl W D, Hassan Y A. Three-dimensional PTV study of the surrounding flow and wake of a bubble rising in a stagnant liquid[J]. Experiments in Fluids.2000, 29(0):202-210.
[52]Xiong Y, Shafer S A. Depth from focusing and defocusing[C],1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, USA,1993:68-67.
[53]Pentland A P. A New sense for depth of field[J]. Pattern Analysis and Machine Intelligence.1987, PAMI-9(4):523-531.
[54]Changyin Z, Cossairt 0, Nayar S. Depth from diffusion[C],2010 IEEE Conference on Computer Vision and Pattern Recognition, New York, USA,2010:1110-1117.
[55]Birch P, Buchanan A, Young R, et al. Depth from structured defocus that is independent of the object reflectivity function[J]. Optical Letters.2011,36(12):2194-2196.
[56]Leroy J V, Simon T, Deschenes F. Real time singular depth from defocus Image and signal processing[J]. Lecture Notes in Computer Science.2008,5099:103-111.
[57]Ben-Ari R, Raveh G. Variational depth from defocus in real-time[C].2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops). Barcelona, Spain,2011: 522-529.
[58]Held R T, Cooper E A, Banks Martin S. Blur and disparity are complementary cues to depth[J]. Current Biology.2012,22(5):426-431.
[59]Subbarao M, Surya G. Depth from defocus: a spatial domain approach[J]. International Journal of Computer Vision.1994,13(3):271-294.
[60]Willert C E, Gharib M. Three-dimensional particle imaging with a single camera[J]. Experiments in Fluids.1992,12(6):353-358.
[61]Schechner Y Y, Kiryati N. Depth from defocus vs. stereo:how different really are they?[J], International Journal of Computer Vision.2000,39(2):141-162.
[62]Baek S J, Lee S J. A new two-frame particle tracking algorithm using match probability[J]. Experiments in Fluids.1996,22(l):23-32.
[63]Pereira F, Ghari M. Defocusing digital particle image velocimetry and the three-dimensional characterization of two-phase flows[J]. Measurement Science and Technology.2002,13(2): 683-669.
[64]Piestun R. Three-dimensional imaging by three-dimensional point spread function encoding[C], Digital Holography and Three-Dimensional Imaging - DH/OTA Joint Session. Vancouver, Canada, 2009:1332-1335.
[65]Huei Y L, Kai D G. Depth recovery using defocus blur at infinity[C].19th International Conference on Pattern Recognition. Tampa, USA,2008:1-4.
[66]Ranipa K R, Joshi M V. A practical approach for depth estimation and image restoration using defocus cue[C].2011 IEEE International Workshop on Machine Learning for Signal Processing. Santander, Philippines,2011:1-6.
[67]Shyh M J. Depth from defocus using radial basis function Networks[C].2007 International Conference on Machine Learning and Cybernetics. Hong Kong, China,2007(4):1888-1893.
[68]Surya G, Subbarao M. Depth from defocus by changing camera aperture:a spatial domain approach[C].1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA,1993:61-67.
[69]Quanbing Z, Yanyan G. A novel technique of image-based camera calibration in depth-from-defocus[C]. First International Conference on Intelligent Networks and Intelligent Systems. WuHan, China,2008:483-486.
[70]Leroy J V, Simon T, Deschenes F. An efficient method for singular depth from defocus [C].50th International Symposium ELMAR-2008. Zadar, Yugoslavia,2008:172:175.
[71]Ding F, Ding J. Least-squares parameter estimation for systems with irregularly missing data[J]. International Journal of Adaptive Control and Signal Processing.2010,24(7):540-553.
[72]Zhuo S, Sim T. On the recovery of depth from a single defocused image. Computer Analysis of Images and Patterns[J].2009,5702(1):889-897.
[73]Pereira F, Gharib M, Modarress D. Defocusing digital particle image velocimetry: a 3-component 3-dimensional DPIV measurement technique. Application to bubbly flows[J]. Experiments in Fluids.2000,29(1):078-084.
[74]Fouras A J D L, Nguyen C V. Volumetric correlation PIV:a new technique for 3D velocimetry vector field measurement[J]. Experiments in Fluids.2009,47(4):569-577.
[75]Hunan H, Dabiri D, Gharib M. On errors of digital particle image velocimetry [J]. Measurement Science and Technology.1997,8(12):1427-1440.
[76]Xiong Y L, Shafer S. Depth from focusing and defocusing (Tech report CMU-RI-93-07)[M] Pennsylvania: Robotics Institute of Carnegie Mellon University,1993.
[77]Trouve P, Champagnat F, Le Besnerais G, et al. Single image local blur identification[C].18th IEEE International Conference on Image Processing. Brussels, Belgium,2011:613-616.
[78]Tomasi C, Manduchi R. Bilateral filtering for gray and color images[C]. Sixth International Conference on Computer Vision. Bombay, India,1998:839-846.
[79]Ming Z, Gunturk B K. Multiresolution bilateral filtering for image denoising[J]. IEEE Transactions on Image Processing.2008,17(12):2324-2333.
[80]Dahyot R. Statistical hough transform[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2009,31(8):1502-1509.
[81]Davies E R. Machine Vision(third Edition):Theory, Algorithms, Practicalities[M]. USA:Elsevier, 2005:284-337.
[82]Rizon M Y H, Saad P. Object detection using circular hough transform[J]. American Journal of Applied Sciences.2005,2(12):1606-1609.
[83]Fujiwara T, Fujimoto K, Maruyama T. A real-time visualization system for PIV[J]. Lecture Notes in Computer Science,2003,2778:437-447.
[84]Davies E R. Circularity - a new principle underlying the design of accurate edge orientation operators[J]. Image and Vision Computing.1984,2(3):134-142.
[85]Jiang X D. Extracting image orientation feature by using integration operator[J]. Pattern Recognition.2007,40(2):705-717.
[86]Fabbri R, Costa L D F, Torelli J C.2D Euclidean distance transform algorithms:A comparative survey[J]. ACM Computer Surveys.2008,40(1):1-44.
[87]Ikoen L. Distance Transform on gray-level surfaces[D], Finland: Lappeenranta University of Technology,2006.
[88]Qu Y D, Cui C S, Chen S B. A fast sub-pixel edge detection method using Sobel-Zernike moments operator[J]. Image and Vision Computing.2005,23(1):11-17.
[89]Norell K, Lindblad J, Svensson S K. Grey weighted polar distance transform for outlining circular and approximately circular objects[C].14th International Conference on Imaging Analysis and Processing. Modena, Italy,2007:647-652.
[90]Kierkegaard P. A method for detection of circular arcs based on the Hough Transform[J]. Machine Vision and Applications.1992,5(5):249-263.
[91]Liu D, Cao Y, Wallapak T. Quadrant coverage histogram: a new method for measuring quality of colonoscopic procedures. Proceedings of the 29th Annual International Conference of the IEEE EMBS. Lyon, France,2007:3470-3473.
[92]Yamashita A, Shirane Y, Kaneko T. Monocular underwater stereo-3D measurement using difference of appearance depending on optical paths[C],2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, China,2010:3652-3657.
[93]Hopkins L M, Kelly J T, Wexler A S, et al. Particle image velocimetry measurements in complex geometries[J]. Experiments in Fluids.2000,29(1):91-95.
[94]Dabiri D. Digital particle image thermometry/velocimetry: a review[J]. Experiments in Fluids. 2009,46(2):191-241.
[95]Hassan Y A, Dominguez-Ontiveros E E. Flow visualization in a pebble bed reactor experiment using PIV and refractive index matching techniques [J]. Nuclear Engineering and Design.2008, 238(11):3080-3085.
[96]Wiederseiner S, Andreini N, Epely-Chauvin G, et al. Refractive-index and density matching in concentrated particle suspensions: a review[J]. Experiments in Fluids.2011,50(5):1183-1206.
[97]Budwig R. Refractive index matching methods for liquid flow investigations[J]. Experiments in Fluids.1994,17(5):350-355.
[98]Prasad A K. Particle image velocimetry measurements in complex geometries [J]. Experiments in Fluids 2000,25(1):91-95.
[99]Wang P, Xiao X, Zhang Z, et al. Study on the position and orientation measurement method with singular vision system[J]. Chinese Optical Letters.2010,8(1):55-58.
[100]Schmalz M S. Stereoscopic imaging through the sea surface:I. theory and error analysis[J]. Proceedings of SPIE.1998,3392(4):134-145.
[101]Arizaga R, Cap N, Rabal H. Image distortion due to refraction by planar surfaces [J]. European Journal of Physics.2010,31(1):115-127.
[102]Treibitz T, Schechner Y Y, Singh H. Flat refractive geometry[C]. IEEE Conference on Computer Vision and Pattern Recognition. Alaska, USA,2009:1-8.
[103]Ricardo P, Mendon a S. Multiview geometry: profiles and self-calibration[D], UK: Department of Engineering of University of Cambridge,2001.
[104]Barta A, Horvth G. Underwater binocular imaging of aerial objects versus the position of eyes relative to the flat water surface[J]. Journal of Optical Society of America.2003, 20(12):2370-2378.
[105]Suiter H. Optics Near the Snell Angle in a Water-to-Air Change of Medium[C]. OCEANS 2007, Panama City, Panama,2007:1-12.
[106]Prasad A K, Jensen K. Scheimpflug stereo camera for particle image velocimetry in liquid flows[J]. Applied Optics.1995,34(30):7092-7099.
[107]Jason G. Refractive epipolar geometry for underwater stereo matching[C].2011 Canadian Conference on Computer and Robot Vision, St. Johns, USA,2011:146-152.
[108]Shimizu M, Okutomi M. Calibration and rectification for reflection stereo[C]. IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA,2008:1-8.
[109]Spinewine B, Capart H, Larcher M, et al. Three-dimensional voronoi imaging methods for the measurement of near-wall particulate flows[J]. Experiments in Fluids.2003,34(2):227-241.
[110]Monica M, Cushman J, Cenedese A. Application of photogrammetric 3D-PTV technique to track particles in porous media[J]. Transport in Porous Media.2009,79(1):43-65.
[111]Stellmacher M, Obermayer K. A new particle tracking algorithm based on deterministic annealing and alternative distance measures[J]. Experiments in Fluids.2000,28(6):506-518.
[112]Panday S P, Ohmi K, Nose K. An ant colony optimization based stereoscopic particle pairing algorithm for three-dimensional particle tracking velocimetry[J]. Flow Measurement and Instrumentation.2011,22(1):86-95.
[113]Labonte G. A new neural network for particle-tracking velocimetry [J]. Experiments in Fluids. 1999,26(4):340-346.
[114]Panday S, Ohmi K, Nose K. An improved ant colony optimization based particle matching algorithm for time-differential pairing in particle tracking velocimetry[J]. Advanced Intelligent Computing Theories and Applications.2010,6216(1):342-349.
[115]Lingfeng X, Au O C, Wenxiu S, et al. Image rectification for single camera stereo system[C], 18th IEEE International Conference on Image Processing, Brussels, Belgium,2011:977-980.
[116]Guoyan X, Lipeng C, Feng G. Study on binocular stereo camera calibration method[C],2011 International Conference on Image Analysis and Signal Processing, Hubei, China,2011:133-137.
[117]Hirschmuller H, Gehrig S. Stereo matching in the presence of sub-pixel calibration errors[C], 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, USA,2009: 437-444.
[118]Bastiaans R JM, van der Plas GAJ, Kieft R N. The performance of a new PTV algorithm applied in super-resolution PIV[J]. Experiments in Fluids.2002,32(3):346-356.
[119]Pereira F, Stuer H, Graff E C, et al. Two-frame 3D particle tracking[J]. Measurement Science and Technology.2006,17(7):1680-1169.
[120]Lee S, Kim S. Advanced particle-based velocimetry techniques for microscale flows[J]. Microfluidics and Nanofluidics.2009,6(5):577-588.
[121]Yong Z, Bin C, Yu F, et al. A parallel implementation of nearest neighbor analysis based on GPU[C],19th International Conference on Geoinformatics, ShangHai, China,2011,2011:1-6.
[122]Chaudhry R, Ivanov Y. Fast approximate nearest neighbor methods for example-based video search[J]. Video Analytics for Business Intelligence.2010,409(1):69-97.
[123]Yihui L, Shuchu X. Feature selection through optimization of k-nearest neighbor matching gain[C],2010 International Conference on Intelligent Computation Technology and Automation, ChangSha, China,2010:309-312.
[124]Dongning L, Yuanhui Z, Yigang S, et al. A multi-frame particle tracking algorithm robust against input noise[J]. Measurement Science and Technology.2008,19(10):105401-105411.
[125]Virant M, Dracos T.3D PTV and its application on Lagrangian motion[J]. Measurement Science and Technology.1997,8(12):1539-1552.
[126]Hai D, Muguo L. The study for particle image velocimetry system based on binocular vision[J]. Measurement.2009,42(4):619-627627.
[127]杜海.三维PIV系统中匹配技术的研究[D],大连：大连理工大学,2009.
[128]王灿星,林建忠,山本富十夫.二维PIV国像处理算法[J].水动力学研究与进展.2001,16(4)：399-404.
[129]Fujiwara T, Fujimoto K, Maruyama T. A real-time visualization system for PIV[J]. Field Programmable Logic and Applications.2003,2778:437-447.
[130]Roth G I, Katz J. Five techniques for increasing the speed and accuracy of PIV interrogation[J]. Measurement Science and Technology.2001,12(1):238-245.
[131]胡广书.数字信号处理(理论、算法与实现)：第二版[M].北京：清华大学出版社,2003：176-177.
[132]Kahler C J, Kompenhans J. Fundamentals of multiple plane stereo particle image velocimetry[J]. Experiments in Fluids.2000,29(0):S070-S077.
[2]Raffel M, Willert C E, Wereley S T, et al. Particle image velocimetry: a practical guide (Experimental Fluid Mechanics)[M]. New York: Springer,2007:8-59.
[3]Maas H G, Gruen A, Papantoniou D. Particle tracking velocimetry in three-dimensional flows[J]. Experiments in Fluids.1993,15(2):133-146.
[4]Arroyo M, Hinsch K. Recent developments of PIV towards 3D measurements Particle Image Velocimetry[M]. Berlin: Springer,2008:127-154.
[5]Haigermoser C, Scarano F, Onorato M. Investigation of the flow in a circular cavity using stereo and tomographic particle image velocimetry [J]. Experiments in Fluids.2009,46(3):517-526.
[6]Coudert S, Foucaut J, Kostas J, et al. Double large field stereoscopic PIV in a high reynolds number turbulent boundary layer[J]. Experiments in Fluids.2011,50(1):1-12.
[7]Berg E J, Robinson R J. Stereoscopic particle image velocimetry analysis of healthy and emphysemic alveolar sac models[J]. Journal of Biomechanical Engineering.2011, 133(6):061004-061008.
[8]Spence C, Buchmann N, Jermy M, et al. Stereoscopic PIV measurements of flow in the nasal cavity with high flow therapy[J]. Experiments in Fluids.2011,50(4):1005-1017.
[9]Willneff J. Spatio-temporal matching algorithm for 3D particle tracking velocimetry[D]. Zurich: Swiss Federal Institute of Technology Zuric,2002.
[10]Kieft R K, Schreel K S, van der Plas GvdP, et al. The application of a 3D PTV algorithm to a mixed convection flow[J]. Experiments in Fluids.2002,33(4):603-611.
[11]Maas H G, Putze T, Westfeld P. Recent developments in 3D-PTV and Tomo-PIV[J]. Imaging Measurement Methods for Flow Analysis.2009,106(1):53-62.
[12]Peterson S D, Chuang H S, Wereley S T. Three-dimensional particle tracking using micro-particle image velocimetry hardware[J]. Measurement Science and Technology.2008, 19(11):1-8.
[13]Peterson K, Regaard B, Heinemann S, et al. Single-camera, three-dimensional particle tracking velocimetry[J]. Optics Express.2012,20(8):9031-9037.
[14]Tarlet D, Bendicks C, Roloff C, et al. Gas flow measurements by 3D particle Tracking velocimetry using coloured tracer particles[J]. Flow, Turbulence and Combustion,2012, 88(3):343-365.
[15]Patel R S, Garimella S V. Development of a particle tracking-based measurement technique to map three-dimensional interfaces between transparent, immiscible fluids[C].13th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems. San Diego, USA,2012:1432-1437.
[16]Ushijima S, Tanaka N. Three-dimensional particle tracking velocimetry with laser-light sheet scanning[J]. Journal of Fluids Engineering.1996,118(2):352-357.
[17]Buchmann N, Atkinson C, Jeremy M, et al. Tomographic particle image velocimetry investigation of the flow in a modeled human carotid artery bifurcation[J]. Experiments in Fluids. 2011,50(4):1131-1151.
[18]Wegfrass A, Lobutova E, Resagk C. Application of 3D particle tracking velocimetry in convection[J]. Springer Proceedings in Physics.2012,141:297-300.
[19]Malek M, Lebrun D, Allano D. Digital in-line holography system for 3D-3C particle tracking velocimetry[J]. Particle Image Velocimetry.2008,112(1):155-170.
[20]Dubsky S, Jamison R, Higgins S, et al. Computed tomographic X-ray velocimetry for simultaneous 3D measurement of velocity and geometry in opaque vessels[J]. Experiments in Fluids.2012,52(3):543-554.
[21]Seifi M, Fournier C, Denis L, et al. Three-dimensional reconstruction of particle holograms:a fast and accurate multiscale approach[J]. Journal of the Optical Society of America.2012, 29(9):1808-1817.
[22]Okamoto K, Schmidl W D, Hassan Y A, Spring model tracking algorithm for three-dimensional particle image velocimetry[J]. ASME Fluids Engineering Division.1995,209:91-97.
[23]Okamoto K, Three-dimensional particle cluster tracking algorithm using affine transformation[J]. Journal of Visualization.1998,1 (2):153-160.
[24]Ishikawa M, Murai Y, Wada A, et al. A novel algorithm for particle tracking velocimetry using the velocity gradient tensor tensor[J]. Experiments in Fluids.2000,29(6):519-531
[25]Ohyama R, Takagi T, Tsukiji T, et al. Particle tracking technique and velocity measurement of visualized flow fields by means of genetic algorithm[J]. Journal of the Visualization Society.1993, 13(S1):35-38.
[26]Kimura I, Hattori A, Ueda M. Particle airing using genetic algorithms for PIV[J]. Journal of Visualization.2000,2(3-4):223-228.
[27]Grant I, Pan X. An investigation of the performance of multi-layer neural networks applied to the analysis of PI V images [J]. Experiments in Fluids.1995,19(3):159-166.
[28]Knaak M. A hopfield neural network for flow field computation based on particle image velocimetry/particle tracking velocimetry image sequences[C], IEEE International Conference on Neural Networks. Houston, USA,1997:48-52.
[29]Ohmi K, Sapkota A. Particle tracking velocimetry using cellular neural networks[C]. International Joint Conference on Neural Networks. Vancouver, Canada,2006:3963-3969.
[30]Labonte G. A new neural network for particle tracking velocimetry[J]. Experiments in Fluids. 1999,26(4):340-346
[31]Wernet M P. Fuzzy logic particle tracking velocimetry[C]. Proceeding of SPIE on Optical Diagnostics in Fluid and Thermal Flow. San Diego, USA,2006:198:200.
[32]Takagi T. Study on particle tracking velocimetry using ant colony optimization [J], Journal of Visualization Society.2007,27(S2):89-90.
[33]Ohmi K, Sapkota A, Panday S P. Applicability of ant colony optimization in particle tracking velocimetry[C].14th International Symposium on Applications of Laser Techniques to Fluid Mechanics. Lisbon, Portugal,2008:1-12.
[34]田晓东,李玉梁,陈嘉范,等.一种新改进的DPIV方法[J].水科学进展.2000,11(1)：59-63.
[35]陈波,李万平.粒子图像测速互相关中亚像素位移定位的改进方法[J].计算机辅助设计与图形学学报.2011,23(11)：1896-1901.
[36]王希麟,张大力,常辙,等.两相流场粒子成像测速技术(PTV-PIV)初探[J].力学学报.1998,30(1)：121-125.
[37]康琦,申功忻.全场测速技术进展[J].力学进展.1997,27(1)：106-121.
[38]许联锋.水气两相流动的数字图像测量方法及应用研究[D],西安：西安理工大学,2004.
[39]白玉川.复杂表面流场中粒子跟踪测速技术的研究[C],第二十届全国水动力学研讨会,太原：《水动力学研究与进展》编委会,2007.
[40]宋连壮.基于图像处理的气固稀相流场测量及气泡行为[D],石家庄：东北电力大学,2011.
[41]周云龙,宋连壮,周红娟.基于特征相似度匹配PTV法的稀相气固两相流流场的测量[J].化工自动化及仪表.2010,37(10)：58-61.
[42]金家莉.二维数字滤波在物理模型流场测量系统中的应用[J].水道港口.2007,28(6)：448-452.
[43]禹明忠.PTV技术和颗粒三维运动规律的研究[D],北京：清华大学,2002.
[44]蔡毅,由长福,祁海鹰,等.模糊逻辑方法用于气固两相流动PTV测量中的颗粒识别过程[J].液体力学实验与测量.2002,16(2)：78-83.
[45]靳斌,杨冠玲,何振江,等.一种利用示踪粒子群体运动特征的PTV方法[J].光学技术.2000,26(1)：16-18.
[46]杜海,李木国.三维粒子测速系统中遗传匹配法的设计[J].系统仿真学报.2008,20(5)：1251-1258.
[47]靳斌,何振江.一种单相机流场速度场三维测量方法[J].光学技术.2000,26(5)：4450450.
[48]罗锐,孙艳飞,杨献勇.应用于微细流场的全息荧光粒子图像三维测速系统[J].工程热物理学报.27(3)：493-495.
[49]吕且妮,葛宝臻,高岩等.乙醇喷雾场粒子尺寸和速度的数字全息测量[J].光子学报.39(2)：266-270.
[50]Hassan Y, Blanchat T. Full-field bubbly flow velocity measurements by digital image pulsed laser velocimetry[J]. Experiments in Fluids.1991,11(5):293-301.
[51]Ortiz-Villafuerte J, Schmidl W D, Hassan Y A. Three-dimensional PTV study of the surrounding flow and wake of a bubble rising in a stagnant liquid[J]. Experiments in Fluids.2000, 29(0):202-210.
[52]Xiong Y, Shafer S A. Depth from focusing and defocusing[C],1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, USA,1993:68-67.
[53]Pentland A P. A New sense for depth of field[J]. Pattern Analysis and Machine Intelligence.1987, PAMI-9(4):523-531.
[54]Changyin Z, Cossairt 0, Nayar S. Depth from diffusion[C],2010 IEEE Conference on Computer Vision and Pattern Recognition, New York, USA,2010:1110-1117.
[55]Birch P, Buchanan A, Young R, et al. Depth from structured defocus that is independent of the object reflectivity function[J]. Optical Letters.2011,36(12):2194-2196.
[56]Leroy J V, Simon T, Deschenes F. Real time singular depth from defocus Image and signal processing[J]. Lecture Notes in Computer Science.2008,5099:103-111.
[57]Ben-Ari R, Raveh G. Variational depth from defocus in real-time[C].2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops). Barcelona, Spain,2011: 522-529.
[58]Held R T, Cooper E A, Banks Martin S. Blur and disparity are complementary cues to depth[J]. Current Biology.2012,22(5):426-431.
[59]Subbarao M, Surya G. Depth from defocus: a spatial domain approach[J]. International Journal of Computer Vision.1994,13(3):271-294.
[60]Willert C E, Gharib M. Three-dimensional particle imaging with a single camera[J]. Experiments in Fluids.1992,12(6):353-358.
[61]Schechner Y Y, Kiryati N. Depth from defocus vs. stereo:how different really are they?[J], International Journal of Computer Vision.2000,39(2):141-162.
[62]Baek S J, Lee S J. A new two-frame particle tracking algorithm using match probability[J]. Experiments in Fluids.1996,22(l):23-32.
[63]Pereira F, Ghari M. Defocusing digital particle image velocimetry and the three-dimensional characterization of two-phase flows[J]. Measurement Science and Technology.2002,13(2): 683-669.
[64]Piestun R. Three-dimensional imaging by three-dimensional point spread function encoding[C], Digital Holography and Three-Dimensional Imaging - DH/OTA Joint Session. Vancouver, Canada, 2009:1332-1335.
[65]Huei Y L, Kai D G. Depth recovery using defocus blur at infinity[C].19th International Conference on Pattern Recognition. Tampa, USA,2008:1-4.
[66]Ranipa K R, Joshi M V. A practical approach for depth estimation and image restoration using defocus cue[C].2011 IEEE International Workshop on Machine Learning for Signal Processing. Santander, Philippines,2011:1-6.
[67]Shyh M J. Depth from defocus using radial basis function Networks[C].2007 International Conference on Machine Learning and Cybernetics. Hong Kong, China,2007(4):1888-1893.
[68]Surya G, Subbarao M. Depth from defocus by changing camera aperture:a spatial domain approach[C].1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA,1993:61-67.
[69]Quanbing Z, Yanyan G. A novel technique of image-based camera calibration in depth-from-defocus[C]. First International Conference on Intelligent Networks and Intelligent Systems. WuHan, China,2008:483-486.
[70]Leroy J V, Simon T, Deschenes F. An efficient method for singular depth from defocus [C].50th International Symposium ELMAR-2008. Zadar, Yugoslavia,2008:172:175.
[71]Ding F, Ding J. Least-squares parameter estimation for systems with irregularly missing data[J]. International Journal of Adaptive Control and Signal Processing.2010,24(7):540-553.
[72]Zhuo S, Sim T. On the recovery of depth from a single defocused image. Computer Analysis of Images and Patterns[J].2009,5702(1):889-897.
[73]Pereira F, Gharib M, Modarress D. Defocusing digital particle image velocimetry: a 3-component 3-dimensional DPIV measurement technique. Application to bubbly flows[J]. Experiments in Fluids.2000,29(1):078-084.
[74]Fouras A J D L, Nguyen C V. Volumetric correlation PIV:a new technique for 3D velocimetry vector field measurement[J]. Experiments in Fluids.2009,47(4):569-577.
[75]Hunan H, Dabiri D, Gharib M. On errors of digital particle image velocimetry [J]. Measurement Science and Technology.1997,8(12):1427-1440.
[76]Xiong Y L, Shafer S. Depth from focusing and defocusing (Tech report CMU-RI-93-07)[M] Pennsylvania: Robotics Institute of Carnegie Mellon University,1993.
[77]Trouve P, Champagnat F, Le Besnerais G, et al. Single image local blur identification[C].18th IEEE International Conference on Image Processing. Brussels, Belgium,2011:613-616.
[78]Tomasi C, Manduchi R. Bilateral filtering for gray and color images[C]. Sixth International Conference on Computer Vision. Bombay, India,1998:839-846.
[79]Ming Z, Gunturk B K. Multiresolution bilateral filtering for image denoising[J]. IEEE Transactions on Image Processing.2008,17(12):2324-2333.
[80]Dahyot R. Statistical hough transform[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence.2009,31(8):1502-1509.
[81]Davies E R. Machine Vision(third Edition):Theory, Algorithms, Practicalities[M]. USA:Elsevier, 2005:284-337.
[82]Rizon M Y H, Saad P. Object detection using circular hough transform[J]. American Journal of Applied Sciences.2005,2(12):1606-1609.
[83]Fujiwara T, Fujimoto K, Maruyama T. A real-time visualization system for PIV[J]. Lecture Notes in Computer Science,2003,2778:437-447.
[84]Davies E R. Circularity - a new principle underlying the design of accurate edge orientation operators[J]. Image and Vision Computing.1984,2(3):134-142.
[85]Jiang X D. Extracting image orientation feature by using integration operator[J]. Pattern Recognition.2007,40(2):705-717.
[86]Fabbri R, Costa L D F, Torelli J C.2D Euclidean distance transform algorithms:A comparative survey[J]. ACM Computer Surveys.2008,40(1):1-44.
[87]Ikoen L. Distance Transform on gray-level surfaces[D], Finland: Lappeenranta University of Technology,2006.
[88]Qu Y D, Cui C S, Chen S B. A fast sub-pixel edge detection method using Sobel-Zernike moments operator[J]. Image and Vision Computing.2005,23(1):11-17.
[89]Norell K, Lindblad J, Svensson S K. Grey weighted polar distance transform for outlining circular and approximately circular objects[C].14th International Conference on Imaging Analysis and Processing. Modena, Italy,2007:647-652.
[90]Kierkegaard P. A method for detection of circular arcs based on the Hough Transform[J]. Machine Vision and Applications.1992,5(5):249-263.
[91]Liu D, Cao Y, Wallapak T. Quadrant coverage histogram: a new method for measuring quality of colonoscopic procedures. Proceedings of the 29th Annual International Conference of the IEEE EMBS. Lyon, France,2007:3470-3473.
[92]Yamashita A, Shirane Y, Kaneko T. Monocular underwater stereo-3D measurement using difference of appearance depending on optical paths[C],2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, China,2010:3652-3657.
[93]Hopkins L M, Kelly J T, Wexler A S, et al. Particle image velocimetry measurements in complex geometries[J]. Experiments in Fluids.2000,29(1):91-95.
[94]Dabiri D. Digital particle image thermometry/velocimetry: a review[J]. Experiments in Fluids. 2009,46(2):191-241.
[95]Hassan Y A, Dominguez-Ontiveros E E. Flow visualization in a pebble bed reactor experiment using PIV and refractive index matching techniques [J]. Nuclear Engineering and Design.2008, 238(11):3080-3085.
[96]Wiederseiner S, Andreini N, Epely-Chauvin G, et al. Refractive-index and density matching in concentrated particle suspensions: a review[J]. Experiments in Fluids.2011,50(5):1183-1206.
[97]Budwig R. Refractive index matching methods for liquid flow investigations[J]. Experiments in Fluids.1994,17(5):350-355.
[98]Prasad A K. Particle image velocimetry measurements in complex geometries [J]. Experiments in Fluids 2000,25(1):91-95.
[99]Wang P, Xiao X, Zhang Z, et al. Study on the position and orientation measurement method with singular vision system[J]. Chinese Optical Letters.2010,8(1):55-58.
[100]Schmalz M S. Stereoscopic imaging through the sea surface:I. theory and error analysis[J]. Proceedings of SPIE.1998,3392(4):134-145.
[101]Arizaga R, Cap N, Rabal H. Image distortion due to refraction by planar surfaces [J]. European Journal of Physics.2010,31(1):115-127.
[102]Treibitz T, Schechner Y Y, Singh H. Flat refractive geometry[C]. IEEE Conference on Computer Vision and Pattern Recognition. Alaska, USA,2009:1-8.
[103]Ricardo P, Mendon a S. Multiview geometry: profiles and self-calibration[D], UK: Department of Engineering of University of Cambridge,2001.
[104]Barta A, Horvth G. Underwater binocular imaging of aerial objects versus the position of eyes relative to the flat water surface[J]. Journal of Optical Society of America.2003, 20(12):2370-2378.
[105]Suiter H. Optics Near the Snell Angle in a Water-to-Air Change of Medium[C]. OCEANS 2007, Panama City, Panama,2007:1-12.
[106]Prasad A K, Jensen K. Scheimpflug stereo camera for particle image velocimetry in liquid flows[J]. Applied Optics.1995,34(30):7092-7099.
[107]Jason G. Refractive epipolar geometry for underwater stereo matching[C].2011 Canadian Conference on Computer and Robot Vision, St. Johns, USA,2011:146-152.
[108]Shimizu M, Okutomi M. Calibration and rectification for reflection stereo[C]. IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA,2008:1-8.
[109]Spinewine B, Capart H, Larcher M, et al. Three-dimensional voronoi imaging methods for the measurement of near-wall particulate flows[J]. Experiments in Fluids.2003,34(2):227-241.
[110]Monica M, Cushman J, Cenedese A. Application of photogrammetric 3D-PTV technique to track particles in porous media[J]. Transport in Porous Media.2009,79(1):43-65.
[111]Stellmacher M, Obermayer K. A new particle tracking algorithm based on deterministic annealing and alternative distance measures[J]. Experiments in Fluids.2000,28(6):506-518.
[112]Panday S P, Ohmi K, Nose K. An ant colony optimization based stereoscopic particle pairing algorithm for three-dimensional particle tracking velocimetry[J]. Flow Measurement and Instrumentation.2011,22(1):86-95.
[113]Labonte G. A new neural network for particle-tracking velocimetry [J]. Experiments in Fluids. 1999,26(4):340-346.
[114]Panday S, Ohmi K, Nose K. An improved ant colony optimization based particle matching algorithm for time-differential pairing in particle tracking velocimetry[J]. Advanced Intelligent Computing Theories and Applications.2010,6216(1):342-349.
[115]Lingfeng X, Au O C, Wenxiu S, et al. Image rectification for single camera stereo system[C], 18th IEEE International Conference on Image Processing, Brussels, Belgium,2011:977-980.
[116]Guoyan X, Lipeng C, Feng G. Study on binocular stereo camera calibration method[C],2011 International Conference on Image Analysis and Signal Processing, Hubei, China,2011:133-137.
[117]Hirschmuller H, Gehrig S. Stereo matching in the presence of sub-pixel calibration errors[C], 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, USA,2009: 437-444.
[118]Bastiaans R JM, van der Plas GAJ, Kieft R N. The performance of a new PTV algorithm applied in super-resolution PIV[J]. Experiments in Fluids.2002,32(3):346-356.
[119]Pereira F, Stuer H, Graff E C, et al. Two-frame 3D particle tracking[J]. Measurement Science and Technology.2006,17(7):1680-1169.
[120]Lee S, Kim S. Advanced particle-based velocimetry techniques for microscale flows[J]. Microfluidics and Nanofluidics.2009,6(5):577-588.
[121]Yong Z, Bin C, Yu F, et al. A parallel implementation of nearest neighbor analysis based on GPU[C],19th International Conference on Geoinformatics, ShangHai, China,2011,2011:1-6.
[122]Chaudhry R, Ivanov Y. Fast approximate nearest neighbor methods for example-based video search[J]. Video Analytics for Business Intelligence.2010,409(1):69-97.
[123]Yihui L, Shuchu X. Feature selection through optimization of k-nearest neighbor matching gain[C],2010 International Conference on Intelligent Computation Technology and Automation, ChangSha, China,2010:309-312.
[124]Dongning L, Yuanhui Z, Yigang S, et al. A multi-frame particle tracking algorithm robust against input noise[J]. Measurement Science and Technology.2008,19(10):105401-105411.
[125]Virant M, Dracos T.3D PTV and its application on Lagrangian motion[J]. Measurement Science and Technology.1997,8(12):1539-1552.
[126]Hai D, Muguo L. The study for particle image velocimetry system based on binocular vision[J]. Measurement.2009,42(4):619-627627.
[127]杜海.三维PIV系统中匹配技术的研究[D],大连：大连理工大学,2009.
[128]王灿星,林建忠,山本富十夫.二维PIV国像处理算法[J].水动力学研究与进展.2001,16(4)：399-404.
[129]Fujiwara T, Fujimoto K, Maruyama T. A real-time visualization system for PIV[J]. Field Programmable Logic and Applications.2003,2778:437-447.
[130]Roth G I, Katz J. Five techniques for increasing the speed and accuracy of PIV interrogation[J]. Measurement Science and Technology.2001,12(1):238-245.
[131]胡广书.数字信号处理(理论、算法与实现)：第二版[M].北京：清华大学出版社,2003：176-177.
[132]Kahler C J, Kompenhans J. Fundamentals of multiple plane stereo particle image velocimetry[J]. Experiments in Fluids.2000,29(0):S070-S077.