基于自适应模拟退火及 LM 联合反演算法的 ECT 图像重建

为解决电容层析成像技术(ECT)中图像重建的非线性和病态性问题,提出了一种自适应模拟退火-Levenberg Marquardt (ASA-LM)联合反演算法。 改进了标准模拟退火(SA)算法的新解生成策略、能量函数的定义及退火策略,并结合 LM 的直接局 部搜索方法联合反演 ECT 图像重建问题。 同时,利用 Savitzky-Golay (SG) 滤波对 ECT 图像重建所需电容数据进行平滑处理以 提高其信噪比。 最后,进行仿真及静态实验,并与线性反投影(LBP)、Landweber 迭代及标准 SA 算法进行了比较。 结果表明,与 其他 3 种算法相比,ASA-LM 算法收敛速度快、图像重建质量明显提高,边缘信息保真度高,重建图像的平均相对误差为 0. 331 1,平均相关系数为 0. 933 1。     

Reconstruction of the projected crystal potential from a periodic high-resolution electron microscopy exit plane wave function

A method based on a simulated annealing algorithm is applied for the reconstruction of the projected crystal potential belonging to a periodic high-resolution electron microscopy exit plane wave function. Using simulated exit plane wave functions of GaAs at different specimen thicknesses, the convergence behaviour and the accuracy of the algorithm are investigated. It is demonstrated that the reconstruction is possible even under strongly non-linear scattering conditions at small specimen thicknesses. Further, the convergence of the algorithm to an ambiguous solution beyond a certain specimen thickness is discussed.     

Reconstruction of pore space from a pore connectivity network via morphological transformations

Pore and grain regions were separated via thresholding techniques from sandstone images. A mathematical morphology-based framework was followed to pack the random pore space with overlapping and nonoverlapping disks. This framework has several advantages in implementation and is generally applicable to multiscale images. The random pore space was reconstructed in two ways from the minimum morphological information through: (a) overlapping and (b) nonoverlapping disks of various shapes and sizes. The structuring elements employed to carry out this analysis included octagon, square and rhombus templates. The results achieved through these two types of reconstruction of sandstone pores are compared. These results provided the basis on which to test the accuracy of these techniques. Reconstruction recovery was tested by computing shapiness indices for the reconstructed pores achieved through the two methods.     

协作制造网络伙伴选择组合模型与算法研究

为满足快速响应市场需求、快速选择和配置资源,提出了三阶段协作制造网络伙伴选择组合优化模型;针对遗传算法早熟收敛和模拟退火算法搜索冗长问题,提出了基于邻域函数的尺度参数自寻优的改进遗传模拟退火算法,并在Delphi环境下通过实验仿真验证了算法的有效性.     

复杂机械产品优化设计实现方法研究

模拟退火算法以随机搜索为基础，以温度足够高为条件，因此收敛速度极为缓慢。为了适应复杂机械产品优化设计，提出一种修正的单纯形法，并用来改进模拟退火算法。该方法既继承了模拟退火方法能避开陷入局部最优的优点，又能搜索到较理想的下降方向，提高了算法的收敛速度。同时提出优化设计的信息模型，构造优化设计的网络关系图，实现了该算法，并成功地应用于抽油机最优化设计。     

Accurate stochastic reconstruction of heterogeneous microstructures by limited x‐ray tomographic projections

An accurate knowledge of the complex microstructure of a heterogeneous material is crucial for its performance prediction, prognosis and optimization. X‐ray tomography has provided a nondestructive means for microstructure characterization in 3D and 4D (i.e. structural evolution over time), in which a material is typically reconstructed from a large number of tomographic projections using filtered‐back‐projection (FBP) method or algebraic reconstruction techniques (ART). Here, we present in detail a stochastic optimization procedure that enables one to accurately reconstruct material microstructure from a small number of absorption contrast x‐ray tomographic projections. This discrete tomography reconstruction procedure is in contrast to the commonly used FBP and ART, which usually requires thousands of projections for accurate microstructure rendition. The utility of our stochastic procedure is first demonstrated by reconstructing a wide class of two‐phase heterogeneous materials including sandstone and hard‐particle packing from simulated limited‐angle projections in both cone‐beam and parallel beam projection geometry. It is then applied to reconstruct tailored Sn‐sphere‐clay‐matrix systems from limited‐angle cone‐beam data obtained via a lab‐scale tomography facility at Arizona State University and parallel‐beam synchrotron data obtained at Advanced Photon Source, Argonne National Laboratory. In addition, we examine the information content of tomography data by successively incorporating larger number of projections and quantifying the accuracy of the reconstructions. We show that only a small number of projections (e.g. 20–40, depending on the complexity of the microstructure of interest and desired resolution) are necessary for accurate material reconstructions via our stochastic procedure, which indicates its high efficiency in using limited structural information. The ramifications of the stochastic reconstruction procedure in 4D materials science are also discussed.     

Point spread function of the general cone-beam x-ray reconstruction formula

In this paper, the spatially varying point spread function (PSF) of the general cone-beam x-ray reconstruction formula is derived. It is proved that the vertical integral of the PSF is a two-dimensional δ function. Therefore, although the general cone-beam formula is not exact in nature, the vertical integral of the reconstructed three-dimensional image is equal to that of the actual image. The band-limited version of the PSF is also derived. The PSF of the general cone-beam formula is simulated, based on which the reconstruction resolution can be approximately inferred.     

Region-of-interest tomography using filtered backprojection: assessing the practical limits

There are many cases where it is desirable to reconstruct at high resolution a small volume from a larger sample. Here we describe the outcomes of a reconstruction trial based on real samples aimed at delineating the practical limits to which a small region of interest can be viewed from a large sample. Our approach has been to artificially truncate the sinograms of whole sample scans to simulate region of interest tomography. A simple filtered back projection algorithm has been applied, with the sinograms extended laterally in a simple manner to make up for the truncated portions. The impact of the degree of truncation (from 0% down to 99%), the number of projections used, as well as the position of the region of interest, on the faithfulness of the reconstruction is evaluated for a range of sample types. We have assessed the nature of, and extent to which, artefacts are introduced and the degree to which simple strategies can minimize them to an acceptable level without the need for complex reconstruction algorithms, projection stitching strategies or very large detectors. It is found that for a wide range of objects the effect of truncation on feature detection is negligible and that excellent images can be reconstructed if the number of projections is calculated not on the basis of the number of pixels on the camera, but on the number of pixels that would be required to scan the whole sample at the chosen pixel resolution. This paper demonstrates that in many cases more sophisticated reconstruction strategies are not necessary.     

Three-dimensional simulation of flow through a porous medium

A three-dimensional simulated random packing of spheres is digitized and progressively densified with dilations of the initial stacking. Flow of fluid through these successive structures is then simulated with geodesic dilations inside the porous phase. The results allow a clear distinction to be made between two factors affecting the flow: the metric properties and the topological properties of the porous network.     

Morphological segmentation of FIB‐SEM data of highly porous media

Nanoporous materials play an important role in modern batteries as well as fuel cells. The materials microstructure needs to be analyzed as it determines the electrochemical properties. However, the microstructure is too fine to be resolved by microcomputed tomography. The method of choice to analyze the microstructure is focused ion beam nanotomography (FIB‐SEM). However, the reconstruction of the porous 3D microstructure from FIB‐SEM image data in general has been an unsolved problem so far. In this paper, we present a new method using morphological operations. First, features are extracted from the data. Subsequently, these features are combined to an initial segmentation, that is then refined by a constrained watershed transformation. We evaluate our method with synthetic data, generated by a simulation of the FIB‐SEM imaging process. We compare the ground truth in the simulated data to the segmentation result. The new method is found to produce a much smaller error than existing techniques.     

一维矩形光栅AFM图像盲探针表面重建模拟研究

探针结构参数的合理选取将直接决定扫描图像及其盲探针修正图像的失真程度。基于此,以一维矩形模拟光栅为典型案例,对该模拟光栅的原子力显微镜(AFM)扫描成像过程与盲探针修正过程进行了仿真,阐明了探针结构参数对扫描成像过程与盲探针修正过程的影响规律。通过建立线宽变化度与半高宽相结合的图像重建误差评价指标,确定了针对该模拟光栅的AFM探针建议结构参数,并取得了良好的光栅图像重建效果。研究表明,应用线宽变化度结合半高宽来综合评价光栅的AFM测量和图像重建过程,有利于提升实际光栅AFM图像盲探针重建的准确度。     

改进后Ray-Box Intersection权系数矩阵的SART+TVM重建算法

针对均匀采样稀疏型重建问题,研究了Ray-Box Intersection和传统的双线性插值两种权系数矩阵计算方法。对Ray-Box Intersection算法的逻辑漏洞进行了改进,利用直线生成法并结合射线斜率和像素位置关系提出了改进算法。该算法利用代数关系求得射线穿过重建区域的起点和终点,然后取不大于其坐标的最大整数,最后采用直线生成法,依据射线斜率沿X或Y方向得到该条射线在投影区域内经过的所有像素块的左下角坐标集合,对于每个像素块,射线与像素块交点距离即为该像素块对射线的加权值。实验结果表明,相比Ray-Box Intersection权系数矩阵算法,改进后算法速度提高了96%,最大峰值信噪比提高了19%;相比双线性插值算法,改进后算法收敛速度更快,在同样时间代价下最大峰值信噪比提高了5%。改进后的Ray-Box Intersection算法能够获得更好的重建图像质量,细节更清晰。     

自适应权值滤波消除图像椒盐噪声的方法

提出了一种自适应权值滤波算法。该方法利用邻域差分法(ROAD)判定滤波中心点的有效性,若判定该点为噪声,则将邻域像素分别赋予不同权值对该点像素重构,邻域像素权重系数的分配随该点有效程度的增加而增大。一方面抑制了图像噪声,另一方面保留了图像细节。实验证明该方法在滤除椒盐噪声时,无论从视觉角度还是数据恢复角度都比常规滤波方法效果有明显改善。当图像中含有35%的椒盐噪声时,利用该方法恢复数据效果比传统中值滤波提高10 dB。     

基于模拟退火遗传算法的PID控制器参数优化

针对遗传算法存在容易早熟的不足,将模拟退火算法融合到遗传算法中,建立了模拟退火遗传算法,并将其应用于PID控制器的参数优化.结果表明,将模拟退火算法融合到遗传算法中是有效的,基于模拟退火遗传算法的PID控制器参数优化是可行的.     

Identification of the three-dimensional gel microstructure from transmission electron micrographs

Mass transport in gels depends crucially on local properties of the gel network. We propose a method for identifying the three‐dimensional (3D) gel microstructure from statistical information in transmission electron micrographs. The gel strand network is modelled as a random graph with nodes and edges (branches). The distribution of edge length, the number of edges at nodes and the angles between edges at a node are estimated from transmission electron micrographs by image analysis methods. The 3D network is simulated by Markov chain Monte Carlo, with a probability function based on the statistical information found from the micrographs. The micrographs are projections of stained gel strands in slices, and we derive a formula for estimating the thickness of the stained gel slice based on the total projected gel strand length and the number of times that gel strands enter or exit the slice.     

浮点数编码改进遗传算法在平面度误差评定中的研究

随着智能制造系统的迅猛发展,应用元启发模式计算方法快速、准确地求解平面度误差值凸显出重大现实意义。为进一步提高平面度误差计算精度,研究了一种基于浮点数编码的改进遗传算法,在原有遗传算法的交叉变异基础之上,引入模拟退火思想,建立最小包容区域法的数学模型,通过计算机仿真获得了最佳适应度收敛曲线和平均适应度收敛曲线,优化结果表明相比传统遗传算法,平面度误差计算精度提高了33.67%。本算法采用浮点数编码、三段式交叉、转轮式选择和最优保存策略,借助模拟退火算法的局部搜索优势,提升了算法的整体性能,且更便于计算机编程,可进一步推广应用到智能测量仪器的其他高精度形位尺寸计算问题领域。     

基于边缘像元投影的微小轴承亚像素边缘检测

对于微小轴承的内径测量,基于机器视觉的非接触测量技术具有广阔的应用前景。针对轴承的边缘像素点,投影将感光像元划分成感光值不同的两个部分,像素单元的最终灰度值为投影两侧局部灰度统计值的面积加权平均值,取该像素的矩形邻域,并对其邻域灰度值进行高斯加权处理,将边缘投影按泰勒公式展开为二次曲线得到高精度亚像素边缘。针对微小轴承进行实验,将本文算法与基于直线拟合的方法进行比较,每毫米包括约217.18个像素,检测的圆度提高0.04%以上,标准差减少不少于7.27%,因此可更准确获取微小轴承内径的亚像素边缘。     

Effect of data incompleteness on the approximation properties of nonparametric estimation of the two-dimensional probability density of independent random variables

Asymptotic properties of a nonparametric estimate of the two-dimensional probability density of independent random variables are studied. Based on these results, the dependence of its efficiency on data incompleteness in the initial statistical information is determined.