ART算法重建图像的影响因素

代数重建法(ART)的图像重建质量和重建时间受到许多因素的影响。采用计算机仿真试验,研究了先验知识、初始值、不同噪声下不同投影数、缺角度等对图像质量及松弛参数选取的影响。试验结果表明,适当地选取初始值及加入先验知识可以极大地提高图像质量,同时,先验知识的加入影响着最佳松弛参数的选取;不同噪声下不同投影数对松弛参数的选取也不同;在缺失一定角度投影的情况下,可以通过均匀地增加投影数来提高图像质量。     

Rotating polar-coordinate ART applied in industrial CT image reconstruction

With incomplete or noisy projections, algebraic reconstruction technique (ART) algorithms can get better image qualities than convolution back-projection (CBP) algorithm. But, it's too slow to meet the practical industrial requirements. Rotating polar-coordinate technique was introduced into ART in this paper, forming Polar-ART (PART) algorithm. PART is 2.7 times as fast as ART with nearly the same quantitative accuracy and it does not need as much memory as ART, which is proved by experiments on ⁶⁰Co container CT inspection system. A drawback of PART is the loss of resolution in reconstructed images.     

工业CT图像重建的ART算法研究

工业CT是一种先进的无损检测手段。传统的滤波反投影算法（FBP）要求完备的投影数据，而实际检测中往往由于客观原因无法获得完全的投影数据。针对不完全投影的数据重建问题，提出了利用代数重建法（ART）进行图像重建。分析了影响ART算法的重建速度和重建质量的因素，通过仿真实验对两种算法的重建质量进行了比较。结果表明，在投影数据不完备时，ART算法的重建质量要优于FBP算法。     

投影数对ART算法重建质量的影响

针对不完全投影数据的重建问题,提出了利用代数重构技术（ART）算法进行图像重建。分析了投影数对重建线性方程组解的影响,在此基础上通过仿真试验具体研究了不同投影角数和不同射线数对重建质量的影响。试验结果表明,增加每个角度下的射线数可有效提高重建质量,在射线数保持不变时,选择合适的投影角数能够获得较好的重建质量。     

基于PSO-OSSART算法的非轴对称电弧发射系数重建质量评价

针对有序子集-联合重建算法(ordered subsets-simultaneous algebraic reconstruction technique, OSSART)重建参数选取主观性强的不足，提出以重建区域误差最小为适应度的随机优化粒子群算法(particle swarm optimization, PSO)来获取最佳重建参数，并对非轴对称电弧发射系数稀疏角度重建质量进行评价. 结果表明，与最大似然函数-期望值最大化算法相比，基于粒子群的OSSART算法不仅能够在大投射角度间隔条件下使重建误差明显降低，而且具有更强的边缘保持能力，能够有效提高电弧中心区域的重建质量.采用OSSART算法，应在180°范围内至少等间距采集6次特征谱线投影，才能保证变极性等离子-熔化极气体保护复合焊(variable polarity plasma arc-metal inert gas, VPPA-MIG复合焊)电弧发射系数场的重建质量.试验结果为非轴对称电弧可靠光谱诊断提供理论依据.     

基于迭代算法的焊接电弧发射系数重建

采用代数迭代重建算法(ART)来恢复焊接电弧等离子体发射系数.开发了ART算法的MATLAB语言计算程序,并通过一个已知的偏置高斯模型验证了该算法的有效性.以一个TIG焊接电弧为例,图像法测量得到电弧光谱强度,利用所开发的ART算法程序重建了发射系数,并将计算结果与常用的Abel逆变换恢复的发射系数进行了比较.结果表明,ART算法的重建精度较高,可用于焊接电弧发射系数的恢复.     

三维图像重建P—FDK算法的一种改进方法

在各种三维锥束重建算法中,近似算法由于数学形式上简单,实现起来容易,而且在锥角比较小的情况下,能够取得较好的重建效果,所以在实际中有着广泛的应用。在各种基于滤波反投影的近似算法中,FDK类型的算法一直是实际应用中的主流。其中P-FDK算法（重新排列投影数据再进行重建）又是FDK算法的一种推广。算法中体素是人为划分的,考虑在实际重建物体时,我们所定义的体素与其相邻的体素有一定的比例关系。因此提出P-FDK算法的三维_六邻接的改进方法,并用采集数据进行重建,试验结果表明,新方法重建的图像边缘比传统P-FDK方法清晰,而且还有抑制噪声的作用。     

基于CUDA的GPU加速代数迭代重建算法

CT迭代重建算法多用于投影数据不完备的情况,但迭代重建算法在普通计算机上的计算非常耗时,主要源于需要反复地进行投影与反投影计算。为此,文章提出了一种基于NVIDIA统一计算设备架构(CUDA)的联合代数重建加速方法。采用基于射线驱动和基于体素驱动的方法分别加速投影与反投影过程。试验结果显示,在不影响重建图像质量的基础上,重建时间大大减少,具有工程应用价值。     

Error analysis of tomographic filters. I: Theory

The technique of computerized tomography is being studied extensively by engineers, physicists and mathematicians to improve the quality of reconstructed images. Certain error estimates are available for the errors occuring in various tomographic algorithms under the assumption that the object cross-section possesses band-limited projection data. It is known, however, that the cross-section function has a finite support, and hence cannot be band-limited. A Sobolev space analysis has already been reported involving certain error estimates for predicting the inherent error in the convolution backprojection algorithm. The present study is an attempt towards developing a simplified two-dimensional Cartesian formula for predicting the comparative performance of the Fourier filters used in the convolution algorithm. This simplified approach involves the Laplacian of the object function and the second-order (Fourier space) derivatives of the filter functions.     

等角扇形束投影的CT重建算法研究

通过线阵探测器CT系统CD-300BX获得C/C复合材料多个角度下的扇形束CT扫描投影数据,设计并实现了滤波反投影算法。讨论了常用滤波函数和内插方式对CT重建质量的影响,采用峰值信噪比和平均梯度对CT图像重建质量进行客观评价,并与CT系统算法重建结果进行对比。研究结果表明:选用R-L滤波和紧邻内插时CT图像重建质量最好,本文开发的滤波反投影算法重建质量达到了CT系统的重建质量,可满足扇形束投影的CT重建需求。     

Artifacts correction method for fan-beam CT with projections asymmetrically truncated on both sides

In fan-beam CT scanning, the scanned field of view is determined by the fan-beam angle and the effective length of a linear array detector. Therefore, when an object extends outside the scanned field of view, CT projection data acquired by the detector will not be complete and is truncated abruptly at the projection boundaries. Furthermore, due to mechanical misalignment of X-ray source, object and the linear array detector, the length of the truncated data on the left side is not equal to its length on the right side. This asymmetrical truncation on both sides will bring out dual bright-band artifacts in the reconstructed images. The purpose of this paper is to develop a method of eliminating these artifacts by projection extension techniques. We first extend the truncated projection unilaterally by using geometrical symmetry property of the fan-beam scanning. Through this method, the projection center of rotation is adjusted to the center of a virtual sinogram and the region of completely scanned field of view is enlarged. The projection values in the extended region equal their equivalent points in the raw sinogram. Thus, values of the supplemented projection data are not approximate and would not decrease the reconstruction accuracy. Next, “mirror extension” technique is preformed to avoid data discontinuity at the edges of a virtual sinogram. Extra data are supplemented to make the projection data on both boundaries attenuate to zero smoothly. Experimental results demonstrated that the dual bright-band artifacts were well eliminated and the correction method could be implemented within the convolution step of a filtered back-projection reconstruction with negligible computational expense.     

Research on the CT image reconstruction of steel tube section from few projections

Most parameters of steel tube can be acquired from the computer tomography (CT) image of its section. However, a large number of projections are needed to reconstruct the image, so the collection and calculation of the projection data consume too much time to inspect steel tube online. For the purpose of solving the problem, we adopt the CT system that can collect the projection data without mechanical scanning motion and propose a modified penalized exponential EM (PEEM) algorithm to reconstruct the steel tube section from only three projections. In the method, the image area is restricted and the pixel value is binarized in order to decrease the number of projections needed and accelerate the convergence speed. At last, the problems of flaw inspection, influence of the restricted area and spatial resolution are discussed. The experiment results show that the accuracy and the reconstruction time of the algorithm meet the requirements of online inspection of steel tube, so it might be possible to be used in practical production.     

基于线阵列探测器的多角度平行投影数据及其图像重建算法

针对工业断层CT进行三维结构成像的应用需求,提出了由多角度的平行投影数据重建三维CT图像的TV-ART迭代算法的新实现方法,其中将Chambolle方法推广至三维情形并用于求解CT图像全变差(TV)最小。使用该方法进行TV求解的重建图像的质量优于基于最速下降法或共轭梯度法的TV-ART迭代算法。此外,该方法具有高度并行性,适合在GPU,FPGA等高速并行计算硬件上实现,从而可以大幅提高图像重建速度。在扫描时间相同的情况下,该方法重建的三维CT图像质量优于已有方法,特别是显著提高了CT图像的轴向分辨率。     

Calculation and application of liquidus projection

1. Introduction Liquidus projection is one of the important types of phase diagrams. Ternary liquidus projection at constant pressure onto a compositional plane is the most commonly used type. It shows the phase rela- tionships between the liquid phase and the other phases. With the isothermal contour lines on the liq- uidus projection, a ternary liquidus projection gives a direct visualization of the liquidus surface. Rhines [1] presented in detail many ternary liquidus projec- tions. It is …     

A projection method for reconstructing X-ray images of arbitrary cross-section

This paper describes a new method for reconstructing cross-sectional images of an object at arbitrary height and arbitrary angled planes. This method, called the geometric projection method, can produce cross-sectional images for arbitrary planes using only those images acquired from the focal plane of an object. This paper develops the mathematical formulae for the image reconstruction of arbitrary height and arbitrary angled planes. Then, by means of the derived formulae, the cross-sectional images of arbitrary planes are realized via computer simulations. Finally, by using experimental results conducted on the solder joints of a ball grid array package, it is shown that cross-sectional images for arbitrary planes of the object can be reconstructed correctly by the geometric projection method.     

Scatter simulation and correction in computed tomography: A reconstruction-integrated approach modelling the forward projection

Scattered photons lower the quality of the reconstructed images in computed tomography. Additionally to scattering in the scintillator material that lowers the resolution, object-scattered photons cause artefacts due to non-linear absorption and are difficult to simulate. In the context of model-based reconstruction we developed a fast approximation for scattered radiation that is integrated in the forward projection of an iterative polychromatic reconstruction method. We develop the modelling of scattered intensities in an iterative maximum-likelihood reconstruction method, demonstrate a comparison with Monte-Carlo simulated scatter-data and show the result of reconstructed data within the field of industrial non-destructive testing.     

基于人工神经网络的电涡流逆问题解

在涡流检测中,常利用最小方差法来进行缺陷重构,这种方法要求正问题的求解效率高。提出一种新的求解正问题方法,该方法利用人工神经网络来估计涡流检测信号,大大提高缺陷重构效率。通过对理想裂纹数值验证显示,该方法可以用数值计算方法得到学习样本对人工神经网络进行训练,并且人工神经网络在学习过程中对噪声不敏感,因而可以有效抑制噪声。     

基于EMD的CT多分辨率重建

提出了基于经验模态分解（EMD）方法进行CT多分辨率重建。该方法利用了EMD方法自适应滤波的特性,以及CT重建过程的线性性质。将投影数据进行EMD分解后,对得到的内蕴模态函数（IMF）层进行特征选择和组合。由于每一IMF分别代表了信号的不同特征尺度,对IMF或IMF的组合进行重建就能直接得到不同的图像特征,从而实现CT多分辨率重建。试验结果表明,基于EMD的CT多分辨率重建,能够在重建的同时直接获取重建图像的多种特征。     

Design of an adaptive line enhancement system using a variable step-size affine-projection algorithm

This paper presents design of an adaptive line enhancement (ALE) system for improving sensor response using a variable step-size affine-projection algorithm (VSS APA). ALE is an adaptive technique that may be used to detect a periodic signal buried in a broadband noise background such as in rotating machinery fault diagnosis. However, most of the conventional methods for ALE system are based primarily on an adaptive filter with the least-mean-square (LMS) error algorithm. Unfortunately, convergence speed is limited when a filtering plant is varied, because the learning process of the adaptive algorithm fails to respond quickly enough to the changing operational conditions. This study proposed a VSS APA for improving both the convergence speed and the performance of the ALE system. Two applications were conducted to compare the performance of the proposed algorithm and various traditional adaptive filtering algorithms. The first application used the proposed ALE system to improve the response of a wheel speed sensor output signal; the other was used for reducing the background noise during rotating machinery fault diagnosis. Both the experimental results indicated that the ALE with VSS APA has an effective performance and convergence for both applications.     

多孔陶瓷锥束CT重建质量优化

利用锥束CT检测系统对多孔陶瓷试样进行扫查与重建,讨论滤波板位置、透照管电压、旋转轴倾斜校正和投影张数等工艺参数对CT重建图像质量的影响。结果表明:在探测器侧进行滤波可以减少试样的散射,提高图像的信噪比和对比度;与85、140 kV透照管电压下的图像相比,透照管电压为130 kV时CT重建的图像噪声低、对比度和清晰度高;对旋转轴进行?1.06°校正可以减少CT重建图像的伪影和几何畸变;CT重建投影数量增加,重建图像的信噪比、对比度和清晰度等随之提高。调整工艺参数可以改善CT重建图像质量,保证多孔陶瓷缺陷检测和微观形貌分析的准确性,拓展CT技术在陶瓷材料检测领域的应用。