基于二维自动主动形状模型的椎间盘核磁共振图像分割算法

针对椎间盘手动建模主观耗时以及现有分割方法不够准确的问题,提出了一种二维自动主动形状模型(2D-AASM)方法,由基于最小描述长度的椎间盘自动统计形状建模、二维局部梯度建模和分割三部分组成。将25组脊柱核磁共振图像(MRI)的椎间盘专家分割结果作为训练集,采用基于最小描述长度的方法确定点对应关系,建立椎间盘T4-5的统计形状模型和二维局部梯度模型,生成形状模型的方差和目标函数值均小于手工和弧长参数方法。模型建立后,通过3组脊柱MRI数据测试提出的分割方法,与传统主动形状模型（ASM）和加入一维局部梯度模型的ASM方法相比,其分割结果具有更高的戴斯系数值,更低的过分割率和欠分割率。实验结果表明,所提方法建立的模型更准确,分割结果更精确。     

基于多尺度MRF的膝关节MRI图像快速分割

膝关节MRI图像中骨骼的精确分割是进一步分割与定量分析膝部软组织的前提。目前膝关节骨骼分割的方法比较耗时或需要一定的人机交互。为解决这一问题,将多尺度MRF方法引入到膝关节MRI分割中,以实现快速无监督的分割。首先建立高斯混合的灰度统计模型,运用MDL准则自动确定类别的数目。建立多尺度MRF的先验模型时,利用尺度间的因果性给出非迭代的计算方法,由细尺度往粗尺度传递统计信息,再由粗尺度往细尺度计算每个像素的最大后验概率,从而实现快速准确的分割。实验结果表明,与单尺度MRF相比,多尺度MRF分割膝关节MRI所需时间大大减少,且精度与专家手动分割标准相当。算法通过建立多尺度马尔可夫随机场模型,完成了低信噪比膝关节MRI图像快速准确分割,可作为进一步自动分割软骨与半月板等软组织的基础。     

图像局部交互熵分割模型的两步快速优化

目的 针对LCK（local correntropy-based K-means）模型收敛速度慢,提出新的基于LCK模型的两步快速分割模型。方法 两步快速分割模型包括粗分割和细分割。1）粗分割：先将待分割的原始图像下采样,减少数据量;然后使用LCK模型对采样后的粗尺度图像进行分割,得到粗分割结果及其相应的粗水平集函数。由于数据量的减少,粗分割步骤可以快速得到近似分割结果。2）细分割：在水平集函数光滑性约束下,将粗分割结果及其对应的粗水平集函数上采样到原始图像的尺度,然后将上采样后的粗水平集函数作为细分割的初始值,利用LCK模型对原始图像进行精细分割。因初始值与真实目标边界很接近,所以只需很少迭代次数就能得到最终分割结果。结果 采用F-score评价方法分析自然以及合成图像的分割结果,并与LCK模型作比较,新的模型F-score数值最大,且迭代次数不大于50。结论 粗分割步骤能在小数据量的情况下,快速分割出粗略的目标;细分割步骤在较好的初始值条件下,能够快速收敛到最终的分割结果,从而有效提高了模型的计算效率和精确性。本文算法主要适用于分割含有未知噪声及灰度非同质的医学图像,且分割效率高。     

结合高斯过程和形变模型的超声图像分割方法

目标区域的先验形状在基于形变模型的超声图像分割方法中扮演着重要的角色。为了提高先验形状模型对目标轮廓形变细节的建模能力,提出了一种基于高斯过程的统计形状模型。目标的形状被表示成一种离散的随机时间序列;利用高斯过程的性质对训练集中的目标形状变化进行统计学习,从而生成目标的先验形状和先验概率。为给形变模型向目标区域的演化提供观测模型,结合超声图像中目标边缘内外灰度变化特征设计了一种径向纹理特征模型。分割的优化被转化为求最大后验概率的过程。基于真实的临床超声图像实验结果显示,与其他方法相比该方法在复杂形变区域和弱边缘区域提供了更准确和鲁棒的结果。     

一种基于Graph Cuts的多尺度乳腺肿块分割方法

提出一种基于Graph Cuts的多尺度乳腺肿块自动分割方法.首先,应用区域统计融合方法对图像进行粗分割,将得到的粗轮廓作为后续Graph Cuts分割的初始轮廓.在迭代优化阶段,引入多尺度分析方法,以高斯金字塔分解得到的多尺度图像序列代替固定尺度的原始图像序列估计高斯混合模型(GMM)参数,将粗糙尺度的易分割性与精细...     

复小波域混合概率图模型的超声医学图像分割

针对存在大量不规则斑点噪声、目标边缘弱化的超声医学图像分割中较难识别目标的问题, 提出了一种复小波域中混合概率图模型的超声医学图像分割算法.采用具有近似平移不变性和良好方向选择性的双树复小波变换(Dual tree-complex wavelet transform, DT-CWT)提取超声医学图像6个方向的高频特征信息; 其次, 为关联目标的弱特征信息并抑制统计独立的高频噪声, 构建了复小波域混合概率图模型; 尺度间"父—子"节点间标记采用贝叶斯网络进行建模, 尺度内邻域间标记采用马尔科夫随机场(Markov random field, MRF)无向图建模, 对复小波域中同尺度的特征系数采用高斯混合模型建模, 尺度内同标记的观测特征采用高斯模型建模; 最后, 用迭代条件模式(Iterated conditional mode, ICM)实现MRF中误分割率最小的能量函数最优解, 获取标记场, 实现超声医学图像分割.实验结果从视觉效果和定量分析两方面验证表明, 本文算法能有效地提取超声图像的弱目标信息, 较好地定位目标区域, 具有较高的分割精度和鲁棒性.     

基于高阶CRF模型的图像语义分割

图像语义分割方法大多基于点对条件随机场模型, 不能定位到单个目标, 并且难以利用全局形状特征, 造成误识。针对这些问题, 提出一种新的高阶条件随机场模型, 将基于全局形状特征的目标检测结果和点对条件随机场模型统一在一个概率模型框架中, 同时完成图像分割、目标检测与识别的任务。利用目标检测器和前背景分割算法获取图像中目标区域, 在目标区域上定义新的高阶能量项。新的高阶条件随机场模型就是高阶能量项和点对条件随机场模型的加权混合模型, 其最优解即为图像语义分割结果。在MSRC-21类数据库上进行的实验验证了该模型能够显著提升图像语义分割性能, 并定位到单个目标。     

有效的SAR图像多尺度分类算法

为提高分类精度,提出一种基于最大期望(EM)与遗传(GA)算法的多尺度SAR图像无监督分类方法.利用多尺度自回归(MAR)模型描述SAR图像中不同尺度之间的统计相依性,提取多尺度特征.应用混合模型描述多尺度特征,并将GA算法与EM算法相结合给出混合模型的参数估计算法,利用最小描述长度(MDL)准则选择模型的分量教.最后使用Bayes分类器实现了图像的分类与分割.该方法集EM算法和GA算法结合后的优点,对设定初值有较少的敏感性,因而避免了局部最优解.应用于SAP图像的实验表明,在分割精度上GA-EM方法优于MAR模型的算法.     

一种新的红外图像快速分割方法研究

研究红外图像目标分割快速优化问题。在高分辨率红外图像的分割中,红外图像存在数据量大、目标的边缘模糊和噪声较大等导致分辨率低和实时性差。为了快速准确分割,提出基于低尺度分割阈值预测的快速红外图像分割方法(LFIRS)。首先建立尺度阶数计算模型以确定保留原始红外图像目标基本信息所需的最小尺度,在分析多尺度过程对具有目标/背景强相关特性的红外图像进行分割阈值,建立多尺度红外图像分割阈值的相关模型(CLM),结合经典二维阈值法得到的最低两个尺度的阈值CLM模型参数,可以通过CLM模型与最小尺度的快速获取原始尺度的分割阈值,实现红外图像的快速分割。实验结果表明,改进方法提高了图像分辨和分割速度,且改善了分割效果。     

基于多尺度统计形状模型的Levelset分割方法

提出并建立了一种基于小波分析的多尺度统计模型，将该统计模型作为先验知识引入Mumford-Shah能量约束函数，从而指导水平集函数进行图像分割。实验表明，当对拓扑结构复杂的医学图像进行分割时，该方法具有明显的效果，同时分割速度和精度都得到了明显改善。     

Efficient shape parameterization method for multidisciplinary global optimization and application to integrated ship hull shape optimization workflow

Multidisciplinary global shape optimization requires a geometric parameterization method that keeps the shape generality while lowering the number of free variables. This paper presents a reduced parameter set parameterization method based on integral B-spline surface capable of both shape and topology variations and suitable for global multidisciplinary optimization. The objective of the paper is to illustrate the advantages of the proposed method in comparison to standard parameterization and to prove that the proposed method can be used in an integrated multidisciplinary workflow. Non-linear fitting is used to test the proposed parameterization performance before the actual optimization. The parameterization method can in this way be tested and pre-selected based on previously existing geometries. Fitting tests were conducted on three shapes with dissimilar geometrical features, and great improvement in shape generality while reducing the number of shape parameters was achieved. The best results are obtained for a small number (up to 50) of optimization variables, where a classical applying of parameterization method requires about two times as many optimization variables to obtain the same fitting capacity.The proposed shape parameterization method was tested in a multidisciplinary ship hull optimization workflow to confirm that it can actually be used in multiobjective optimization problems. The workflow integrates shape parameterization with hydrodynamic, structural and geometry analysis tools. In comparison to classical local and global optimization methods, the evolutionary algorithm allows for fully autonomous design with an ability to generate a wide Pareto front without a need for an initial solution.     

基于参数化求和不变量与特征重整的形状匹配

从特征提取和特征匹配两方面考虑,提出了一种鲁棒的形状匹配方法。首先,基于求和不变量,设计了基于面积的形状参数化和归一化方法,提出了参数化求和不变量,该不变量基于形状局部描述且采用积分算子计算,具有较好的鲁棒性和仿射不变性。然后,为进一步提高形状匹配的鲁棒性,在特征匹配上,分析了参数化求和不变量的先验信息,设计了基于特征重整的匹配距离函数,并通过动态规划进行实现。仿真实验表明了所提方法的有效性。     

Para-Graph: Graph-Based Parameterization of Triangle Meshes with Arbitrary Genus

This paper describes a novel approach to the parameterization of triangle meshes representing 2‐manifolds with an arbitrary genus. A topology‐based decomposition of the shape is computed and used to segment the shape into primitives, which define a chart decomposition of the mesh. Then, each chart is parameterized using an extension of the barycentric coordinates method. The charts are all 0‐genus and can be of three types only, depending on the number of boundary components. The chart decomposition and the parameterization are used to define a shape graph where each node represents one primitive and the arcs code the adjacency relationships between the primitives. Conical and cylindrical primitives are coded together with their skeletal lines that are computed from and aligned with their parameterization. The application of the parameterization approach to remeshing guarantees that extraordinary vertices are localized only where two patches share a boundary and they are not scattered on the whole surface.     

A New Method Of Curve Parameterization with Applications in Computer Aided Design

A new method of parameterization, based on areas, is suggested and some results obtained for planar cubic curves are presented and compared with standard methods. The new technique provides a "tight" interpolant, is axis independent and offers shape control via a set of off geometry data points (henceforth called poles) relative to which areas are computed. A simple algorithm for the computation of initial pole positions is given (this is based on obtaining agreement with a standard parameterized curve on the same data and with identical geometrical end conditions). Some additional properties of the parameterization are discussed including its limiting behaviour and its applicability to function drawing.     

三角网格曲面共形参数化研究综述

曲面参数化是计算机图形学和计算机辅助设计中的一个重要问题,有着广泛的应用背景。曲面 参数化旨在谋求从三维曲面到某参数区域变换过程中某些内蕴几何量(例如边长、角度、面积)的变形最小化。 保角参数化前后保持曲面局部形状,又称作共形参数化。本文针对三角网格曲面,介绍了共形参数化领域的研 究现状。根据目标参数域的不同,主要讨论了平面参数化、球面参数化和双曲空间参数化。最后总结了现阶段 共形参数化存在的难点问题和未来可能的研究方向。     

High-fidelity aerostructural optimization with integrated geometry parameterization and mesh movement

This paper extends an integrated geometry parameterization and mesh movement strategy for aerodynamic shape optimization to high-fidelity aerostructural optimization based on steady analysis. This approach provides an analytical geometry representation while enabling efficient mesh movement even for very large shape changes, thus facilitating efficient and robust aerostructural optimization. The geometry parameterization methodology uses B-spline surface patches to describe the undeflected design and flying shapes with a compact yet flexible set of parameters. The geometries represented are therefore independent of the mesh used for the flow analysis, which is an important advantage to this approach. The geometry parameterization is integrated with an efficient and robust grid movement algorithm which operates on a set of B-spline volumes that parameterize and control the flow grid. A simple technique is introduced to translate the shape changes described by the geometry parameterization to the internal structure. A three-field formulation of the discrete aerostructural residual is adopted, coupling the mesh movement equations with the discretized three-dimensional inviscid flow equations, as well as a linear structural analysis. Gradients needed for optimization are computed with a three-field coupled adjoint approach. Capabilities of the framework are demonstrated via a number of applications involving substantial geometric changes.     

Visualizing shape transformation between chimpanzee and human braincases

The quantitative comparison of the form of the braincase is a central issue in paleoanthropology (i.e., the study of human evolution based on fossil evidence). The major difficulty is that there are only few locations defining biological correspondence between individual braincases. In this paper, we use mesh parameterization techniques to tackle this problem. We propose a method to conformally parameterize the genus-0 surface of the braincase on the sphere and to calibrate the parameterization to match biological constraints. The resulting consistent parameterization gives detailed information about shape differences between the braincase of human and chimp. This opens up new perspectives for the quantitative comparison of “featureless” biological structures.     

Mesh parameterization based on one-step inverse forming

A novel planar mesh parameterization algorithm via flattening or unfolding is proposed by introducing a so-called one-step inverse approach (IA) based on physical plastic deformation of metal materials. As opposed to methods based on geometric idea, the algorithm is more suitable for CAD models, e.g. rapid prediction of blank shape in sheet metal forming for auto-body panels. Benefiting from proper pre-processing steps, rapid simulation of sheet metal forming and optimal initial solution guess, the proposed algorithm flattens meshes onto a plane with lower area distortion and minimizes the angular distortion. Some numerical examples of results generated by the new parameterization method are provided, and the corresponding analyses are given as well, such as distributions of strain, stress, thickness distribution and formability, etc.     

有理Bezier曲线参数化方法研究

研究了在曲线形状保持不变的条件下,有理n次Bézier曲线的权因子改变与曲线参数化的关系。同时,给出了有理n次Bézier曲线上点的参数与权因子之间的对应关系,导出了有理n次Bézer曲线的n-1个形状不变因子。得到了与权因子变换对参数化有同样影响的参数射影变换,两种变换都不改变曲线的形状和首末端点,仅仅改变了曲线上的点与定义域内点的对应关系。