医学图像的交互分割及三维表面重建

提出一种基于灰度信息的三维医学图像交互分割方法,并对分割提取的组织或区域进行了三维表面重建,在分割过程中,首先对图像进行滤波,确定阈值范围进行二值化,然后根据待提取组织或区域的特征,选取合适的数学形态学操作,最后对区域进行种子填充,这一过程是一个交互的过程,对提取出的区域用MT算法构造等值面,用灰度梯度法计算表面法微量,并实现高质量三维显示。     

基于VTK的医学三维图像模型构建与切割

虚拟外科手术是医学领域的一个重要研究方向。基于面绘制和体绘制两种可视化算法实现医学图像的三维模型重建,针对不同重建模型分别实现虚拟任意切割：面切割和体切割,并讨论了虚拟切割的交互操作实现方法。用VTK（Visualization Toolkit）初步实现了模拟手术系统中的虚拟切割脊柱体功能。     

基于图像的启发式三维火焰重建算法

研究了火焰的重建,为改进现有稀疏视角输入下的层析成像算法的重建精度,提出了一种基于图像的启发式火焰重建方法。该重建方法通过迭代优化求解能量约束模型,实现对火焰三维温度场的重建。算法允许复杂的成像模型应用于重建过程中,使渲染过程更接近火焰真实成像过程,并对数据平滑性进行约束,以提高重建结果质量。算法在迭代过程中使用启发式规则引导温度场数据调整,使算法快速收敛,并通过使用图形处理器加速重建过程。使用模拟数据和真实捕获数据进行实验,结果表明,与现有方法相比,提出的方法有效提高了重建结果的精度及平滑性,验证了提出方法的有效性。     

三维医学图像优化压缩算法

针对 CT 三维医学图像特点,将三维医学数据分块,通过平滑处理和边界交迭减小分块失真,将 SPIHT 算法与 SPECK 算法结合,帧内采用 SPIHT 算法,帧间利用 SPECK 编码算法,组合算法提高了压缩效率,优于单一的算法。     

多平台的三维医学CAD模型重建技术研究

与二维医学图像相比,三维医学模型在解剖结构观测、辅助手术、制造支持等方面具有较大优势.由于三维医学CAD模型重建的复杂性,存在理论研究较多,实际应用不足的情况.针对这一问题,提出了以医学图像处理技术和反求工程技术为基础、结合正向设计技术的三维医学模型的跨平台重建策略.实际应用表明,该策略可方便有效地重建复杂的医学模型.     

列车车轮三维结构光检测中的点云处理研究

列车轮对作为转向架的关键零部件,其检测手段仍以人工检测为主,现有的自动检测方案,大多针对车轮某一断面的参数尺寸进行测量,难以真实反映车轮轮缘踏面的损伤情况。为此,该文提出一种列车车轮三维结构光检测中的点云处理方案。首先,利用三维结构光测量仪器采集列车车轮的三维点云数据;其次,根据列车车轮三维点云的特点,确定包括离群点去除、点云配准、点云平滑处理以及孔洞修补在内的点云处理方案,并对各处理步骤的最优参数进行分析;最后,利用贪婪投影三角化算法,进行列车车轮三维点云数据的曲面重建,使用拉普拉斯平滑算法对重建后的曲面进行平滑处理。结果表明,该文所提出的列车车轮点云处理方案能够实现对三维点云数据的处理,最终得到的列车车轮的三维曲面模型与基准模型的标准偏差为1.768 mm,实现对于列车车轮的三维检测。     

台风作用下风-浪-潮三维环境等值面模型

台风登陆时常伴随强风、巨浪和风暴潮,如何获取台风下的风、浪、潮数据并确定给定重现期下的环境参数成为制约跨海桥梁设计的关键科学问题。以跨海桥梁众多的舟山海域为例,采用SWAN+ADCIRC耦合模型,对过去32年间影响该海域的49次台风过程依次开展数值模拟,利用模拟出的台风作用下风、浪、潮参数建立100年重现期下的环境等值面,讨论了环境参数选取方法对环境等值面模型的影响。研究表明:该文建立的数值模型能为后期构建联合概率及环境等值面模型提供可靠的数据来源;三维嵌套Frank Copula能较好拟合风-浪-潮三维联合概率分布;给定重现期的环境等值面形状受数据选取时刻影响较大,其中风暴潮增水最大值对选取方法最为敏感。     

采用标记拼接的相位测量轮廓术

提出一种采用标记拼接的新的光学三维传感方法。在采用结构照明的光学三维传感中,阴影及条纹错位关系到被测量物体三维面形重建的质量。在处理阴影时,传统方法是采用插值方法,在处理条纹错位时,通常采用增大条纹周期或采用几组条纹周期不同的光波进行投影,这样通常降低了测量精度。为了提高测量精度,本文提出一种在同一参考面内,采用单组相移正弦条纹进行投影,通过旋转被测物体,并对旋转前后两次重建物体在信息可靠区域内进行标记拼接,恢复物体表面完整三维信息的方法。实验证实了此方法的有效性及实用性。     

基于点云的球面三维逆向建模

球面拟合是三维逆向建模中面临的亟待解决的复杂难题之一。它广泛应用于零件检测、建筑物结构恢复建模、医学血管和细胞模拟领域。论文改进了对三维球面点云进行直接拟合的方法,得到球面的相关几何参量,并在速度和精确度上都获得了提高。论文对拟合相关参数和噪声影响进行分析,并否定了对选定的数据集分组整合的算法。LM(Levenberg-Marquardt)算法是最广泛应用的最小二乘法二次曲面拟合的方法之一,本算法在与LM算法的对比中凸显了算法在时间和某些情况下精度的优越性,为进一步研究三维图形拟合重建恢复等工作打下了基础。     

主动激光散斑投射的面部测量方法研究

针对医学等相关领域对面部轮廓测量的需求,提出一种主动激光散斑投射的面部点云获取方法。该方法结合数字图像相关法和双目立体视觉原理,能够快速重建面部三维轮廓。首先,介绍数字图像相关法的基本原理;其次,提出一种匹配前的散斑图像预处理方法,去除散斑图像的边缘化区域,提高散斑匹配质量;最后,计算面部三维点云。为验证该方法的可行性及最终重建效果,设计面部扫描重建及误差实验,结果表明:该方法具有体积小和不受光强影响的特点,能够较好地重建出面部模型,整体测量误差为±0.2 mm。     

A fast algorithm for multifield representation and multiscale simulation of high‐quality 3D stochastic aggregate microstructures by concurrent coupling of stationary Gaussian and fractional Brownian random fields

This study presents a multiscale computational framework for the representation and generation of concrete aggregate microstructures on the basis of the multifield theory, which couples the stationary Gaussian random field with the fractional Brownian random field. Specifically, the stationary Gaussian field is utilized to simulate the morphological shape of an aggregate on the coarse scale, whereas the surface topography of the aggregate on the fine scale is represented by the fractional Brownian field. To bridge the 2 scales, a concurrent coupling formula is proposed. This coupled technique allows for smooth transition between the coarse and fine scales and permits the rapid generation of highly realistic concrete aggregates that can be tailored to the desired quality and requirements, making the algorithm computationally appealing. In the generation of the random fields on the 2 scales, the Fourier representation of block circulant covariance matrices with circulant blocks is exploited, which yields substantial efficiency advantages over the conventional Cholesky decomposition approach in factorizing covariance matrices as well as simulating random fields. Meanwhile, a microsurface postprocessing and reconstruction procedure is also developed to convert the generated random fields into realistic 3D shapes. The numerical methodology proposed in this study offers tremendous potential for a plethora of applications in cement‐based materials.     

A New Encryption-then-Compression Scheme on Gray Images Using the Markov Random Field

Compressing encrypted images remains a challenge. As illustrated in our previous work on compression of encrypted binary images, it is preferable to exploit statistical characteristics at the receiver. Through this line, we characterize statistical correlations between adjacent bitplanes of a gray image with the Markov random field (MRF), represent it with a factor graph, and integrate the constructed MRF factor graph in that for binary image reconstruction, which gives rise to a joint factor graph for gray images reconstruction (JFGIR). By exploiting the JFGIR at the receiver to facilitate the reconstruction of the original bitplanes and deriving theoretically the sum-product algorithm (SPA) adapted to the JFGIR, a novel MRF-based encryption-then-compression (ETC) scheme is thus proposed. After preferable universal parameters of the MRF between adjacent bitplanes are sought via a numerical manner, extensive experimental simulations are then carried out to show that the proposed scheme successfully compresses the first 3 and 4 most significant bitplanes (MSBs) for most test gray images and the others with a large portion of smooth area, respectively. Thus, the proposed scheme achieves significant improvement against the state-of-the-art leveraging the 2-D Markov source model at the receiver and is comparable or somewhat inferior to that using the resolution-progressive strategy in recovery.     

Theory and algorithms for image reconstruction on chords and within regions of interest

We introduce a formula for image reconstruction on a chord of a general source trajectory. We subsequently develop three algorithms for exact image reconstruction on a chord from data acquired with the general trajectory. Interestingly, two of the developed algorithms can accommodate data containing transverse truncations. The widely used helical trajectory and other trajectories discussed in literature can be interpreted as special cases of the general trajectory, and the developed theory and algorithms are thus directly applicable to reconstructing images exactly from data acquired with these trajectories. For instance, chords on a helical trajectory are equivalent to the n-PI-line segments. In this situation, the proposed algorithms become the algorithms that we proposed previously for image reconstruction on PI-line segments. We have performed preliminary numerical studies, which include the study on image reconstruction on chords of two-circle trajectory, which is nonsmooth, and on n-PI lines of a helical trajectory, which is smooth. Quantitative results of these studies verify and demonstrate the proposed theory and algorithms.     

Parallel computing of a digital hologram and particle searching for microdigital-holographic particle-tracking velocimetry

We have developed a parallel algorithm for microdigital-holographic particle-tracking velocimetry. The algorithm is used in (1) numerical reconstruction of a particle image computer using a digital hologram, and (2) searching for particles. The numerical reconstruction from the digital hologram makes use of the Fresnel diffraction equation and the FFT (fast Fourier transform), whereas the particle search algorithm looks for local maximum graduation in a reconstruction field represented by a 3D matrix. To achieve high performance computing for both calculations (reconstruction and particle search), two memory partitions are allocated to the 3D matrix. In this matrix, the reconstruction part consists of horizontally placed 2D memory partitions on the x-y plane for the FFT, whereas, the particle search part consists of vertically placed 2D memory partitions set along the z axes. Consequently, the scalability can be obtained for the proportion of processor elements, where the benchmarks are carried out for parallel computation by a SGI Altix machine.     

Smooth C1‐interpolations for two‐dimensional frictional contact problems

Finite deformation contact problems are associated with large sliding in the contact area. Thus, in the discrete problem a slave node can slide over several master segments. Standard contact formulations of surfaces discretized by low order finite elements leads to sudden changes in the surface normal field. This can cause loss of convergence properties in the solution procedure and furthermore may initiate jumps in the velocity field in dynamic solutions. Furthermore non‐smooth contact discretizations can lead to incorrect results in special cases where a good approximation of the contacting surfaces is needed. In this paper a smooth contact discretization is developed which circumvents most of the aformentioned problems. A smooth deformed surface with no slope discontinuities between segments is obtained by a C¹‐continuous interpolation of the master surface. Different forms of discretizations are possible. Among these are Bézier, Hermitian or other types of spline interpolations. In this paper we compare two formulations which can be used to obtain smooth normal and tangent fields for frictional contact of deformable bodies. The formulation is developed for two‐dimensional applications and includes finite deformation behaviour. Examples show the performance of the new discretization technique for contact. Copyright © 2001 John Wiley & Sons, Ltd.     

圆柱壳辐射声场重构的波叠加方法

将声全息技术应用到水下结构的声场重建问题中。通过理论推导和数值仿真,分析一两端带半球帽的圆柱壳模型在不同激励作用下声场的重建效果及影响重建精度的因素。考虑到实际应用,全息面分别采用柱形全息面和平行的双平面全息面。仿真结果表明,该方法是一种稳健的全波数空间声场重构技术;其重建精度受到等效源面参数、全息面参数及测量环境信噪比等多因素的影响;在含有测量噪声的条件下,应用Tikhonov结合L曲线的正则化方法仍可以较精确地重构声场;相比于其它方法,波叠加方法具有测点少,计算速度快等优点,有很好的应用前景。     

Continuous and discrete energy distributions of electrons field-emitted from carbon nanotubes

Carbon nanotubes (CNTs) deposited by means of electrophoresis onto a tungsten foil have been studied using field electron microscope and a dispersive electron energy analyzer. The field-emitted electron energy distributions (FEEDs) were either normal (smooth) or exhibited a system of narrow peaks spaced by 30 and 55 meV, which were retained after thermal and field-induced reconstruction of the CTN vertex. The observed features of the FEEDs are consistent with a quantum-dimensional model of field electron emission from CNTs.     

Superresolution planar diffraction tomography through evanescent fields

We consider the problem of noninvasively locating objects buried in a layered medium such as land mines in the ground or objects concealed in a wall. In such environments, the transmitter(s) and receiver(s) are frequently within the near‐field region of the illuminating radiation. In these cases, the scattered evanescent field carries useful information on the scattering object. Conventional diffraction tomography techniques neglect, by their design, the evanescent field. Under near‐field conditions, they treat it as noise as opposed to valid data. If correctly incorporated into a reconstruction algorithm, the evanescent field, which carries high spatial frequency information, can be used to achieve resolution beyond the classical limit of λ/2, or “superresolution.” We build on the generalized holography theory presented by Langenberg to develop a planar diffraction tomography algorithm that incorporates evanescent field information to achieve superresolution. Our theory is based on a generalization of the Fourier transform, which allows for complex spatial frequencies in a manner similar to the Laplace transform. We specialize our model to the case of a two‐dimensional multimonostatic, wideband imaging system, and derive an extended resolution reconstruction procedure. We implement and apply our reconstruction to two data sets collected using the Lawrence Livermore National Laboratory (LLNL) Micropower Impulse Radar (MIR). © 2002 John Wiley & Sons, Inc. Int J Imaging Syst Technol 12, 16–26, 2002     

Comparison of Dynamic Differential Evolution and Self-Adaptive Dynamic Differential Evolution for Buried Metallic Cylinder

The application of two techniques for the reconstruction of shape reconstruction of a metallic cylinder from scattered field measurements is studied in this paper. These techniques are applied to two-dimensional configurations, for which the method of moment (MoM) is applied to solve the integral equations. Considering that the microwave imaging is recast as a nonlinear optimization problem, an objective function is defined by the norm of the difference between the measured scattered electric fields and those calculated for each estimated metallic cylinder. Thus, the shape of a metallic cylinder can be obtained by minimizing the objective function. In order to solve this inverse scattering problem, two techniques are employed. The first one is based on dynamic differential evolution (DDE) algorithm, while the second one is an improved version of the DDE algorithm with self-adaptive control parameters, called SADDE. Both techniques are tested for the simulated data contaminated by additive white Gaussian noise. Numerical results indicate that SADDE algorithm outperforms DDE algorithm in terms of reconstruction accuracy and convergence speed.     

Partially adhesive fixation of reconstruction plates at midfacial fractures – an alternative solution to screw fixation?

For the prognosis and long‐term stability of an implant, for example a hip‐ or knee‐implant, the anchorage to the bone is of critical importance. Normally for this anchorage bone cement is used, which achieves sufficient retention to spongious bone structure. If there is the need for osteosynthesis of midfacial fractures for surgical therapy, in some situations it would be clinically preferable to fix the reconstruction plates without the usual system using screws. This is the aim of the new adhesive fixing technology which is presented in this article. The structure of bone in midface, however, is cortical and such a smooth surface will allow only insufficient retention forces between bone and bone cement. In order to obtain a good adhesion at cortical bone nevertheless, we have to apply a bone bonding agent similar to techniques used in dentistry, which are based on dentin bonding agents, because there are different wetting properties of the hydrophilic bone and the hydrophobic bone cement monomers. Determination of achievable bond strength was done in a tension test ex vivo. Reconstruction plates were fixed with bone cement at bovine bone conditioned with a bone bonding agent developed for this field of application. With bond strengths up to 6 MPa it is assumed that an adhesive fixation of reconstruction plates at cortical bone structure is possible using the developed amphiphilic bone bonding agent.