A combined nonlinear contrast image reconstruction algorithm under inexact point-spread function

The problem of image reconstruction is considered for the case when the right-hand side of the 2D integral equation and the point-spread function (the integral equation kernel) are given with random errors. A stable image reconstruction algorithm is proposed. It is a combination of a regularizing algorithm for solving an integral equation (frequency filtering) and a local nonlinear filter (spatial filtering). Characteristics of the 2D point-spread function of the regularizing algorithm are introduced. The regularization parameter is chosen according to the required regularizing algorithm resolution. For eliminating the random reconstruction error, the regularized solution is subjected to nonlinear local filtering that preserves high-frequency information components of the image.     

基于两步迭代TV正则化的电阻抗图像重建算法

针对电阻抗层析成像(electrical impedance tomography,EIT)逆问题求解的欠定性和病态性,克服传统基于L2范数的Tikhonov正则化对介质边界的模糊效应,提出一种基于两步迭代的正则化图像重建算法.该算法采用具有良好保边性的总变差(total variation,TV)正则化函数,利用两步迭代法引入TV去噪算子,达到解的双重正则化效果.与传统最小二乘迭代算法、TV相关迭代算法相比,不仅保证了逆问题求解的稳定性,而且进一步提高了非连续分布介质区域成像的分辨能力,具有较好的成像精度.     

A nonlinear fuzzy assisted image reconstruction algorithm for electrical capacitance tomography

A nonlinear method based on a Fuzzy Inference System (FIS) to improve the images obtained from Electrical Capacitance Tomography (ECT) is proposed. Estimation of the molten metal characteristic in the Lost Foam Casting (LFC) process is a novel application in the area of the tomography process. The convergence rate of iterative image reconstruction techniques is dependent on the accuracy of the first image. The possibility of the existence of metal in the first image is computed by the proposed fuzzy system. This first image is passed to an iterative image reconstruction technique to get more precise images and to speed up the convergence rate. The proposed technique is able to detect the position of the metal on the periphery of the imaging area by using just eight capacitive sensors. The final results demonstrate the advantage of using the FIS compared to the performance of the iterative back projection image reconstruction technique.     

用于电容层析成像的一步稳定图像重建算法

本文提出一种基于Landweber迭代的电容层析成像系统一步稳定图像重建算法。该算法通过构建压缩算子可以克服传统Landweber迭代方法存在的半收敛性,得到稳定的重建图像。同时该算法的矩阵算子可以预先计算,然后通过一步矩阵运算实现图像重构。实验结果表明,该算法可保证重建图像的实时性和鲁棒性。     

PSO算法优化BP神经网络的EIT图像重建算法

针对阻抗断层图像重建这个严重病态的非线性的逆问题,提出一种基于PSO优化BP神经网络的EIT图像重建算法。该算法在基本BP算法的误差反向传播调整权值的基础上,再引入PSO算法进行权值修正。该算法不仅能很好地适应EIT的病态非线性特性,而且可以克服基本BP算法收敛速度慢和易陷于局部极值的局限。实验结果表明该方法速度快并且能够有效提高图像分辨率。     

锥束ART算法快速图像重建

针对锥束ART算法重建速度慢的问题,提出了一种基于投影的三维射线与体素的快速遍历和求交算法.该算法将三维射线投影到两个互相垂直的平面上,通过计算投影线与投影平面的相交情况来确定三维射线穿过体素的索引及长度.利用该算法在图像重建过程中实时计算权因子,不仅节省了大量的内存空间,而且提高了投影和反投影运算的速度.基于该算法的特点,在重建过程中采用了一种按列优先的策略,减少了不必要的计算,大大提高了重建速度.仿真实验表明,该算法非常有效,与传统的Siddon算法相比取得了17倍以上的重建加速比.     

电磁层析成像图像重建中的修正共轭梯度算法

通过研究共轭梯度算法,推导出适用于电磁层析成像的修正共轭梯度算法,该方法提高了收敛速度,改善了电磁层析成像重建图像的质量。首先以共轭搜索方向充分下降为充分条件,理论推导出修正共轭梯度算法。然后从相对图像误差、相关系数和收敛曲线几个方面出发,评价了Landweber迭代法、单步Tikhonov正则化方法、共轭梯度法和修正共轭梯度法在电磁层析成像图像重建中的结果,得出结论:修正共轭梯度方法的相对图像误差最小,重建图像和原图像的相关系数最高,收敛情况优于共轭梯度算法。     

基于LU分解的共轭梯度法单步成像算法

单步和迭代电容层析成像（ECT）图像重建算法分别具有成像质量差和成像速度慢的缺点,为了快速得到高质量的重建图像,本文提出了一种新型的基于LU分解的单步共轭梯度成像法。该方法首先将ECT物理模型进行规范化和Tikhonov正则化处理,进而将LU分解的思想引入ECT方程的求解中,从而实现了单步图像重建。理论分析表明,该算法具有良好的单步收敛性。通过典型流型的仿真实验,证明了该算法可以获得优于反投影算法的重建图像。     

基于神经网络的图像高反差算法的研究与实现

研究了图像增强中的高反差算法,提出将BP神经网络用于实现图像增强中的高反差处理算法,将图像增强转化为参数的优化,可用于实时的图像处理.实验结果表明,该方法是一种有效的图像增强算法.对该算法的仿真实现过程进行了描述,说明了该算法的实用性.     

基于时间串流的 OSEM 图像重建算法研究

利用γ光子探测腔体内部动态流场需要快速的图像重建算法，传统处理方式是先采集所有事件、再进行OSEM等算法处理。本文提出了一种按时间流对响应事件进行子集划分的图像重建(T-OSEM)算法。在连续采样数据的同时，按时间段将采样到的数据划分为子采样数据集，对子集进行OSEM迭代实现图像重建。并将上一帧图像作为迭代输入，利用帧间图像相关性，以加快收敛速度。该算法中数据流的采样与上一帧图像的处理同时进行，并通过多线程并行运算加速图像重建过程。研究了最优子集事件数量及相对应采样时间的关系，以实现在尽可能短的采样时间下达到最优的重建效果。实验表明，当采样时间段达到1 s时，T-OSEM算法仍有很好的粒子跟踪效果，粒子轨迹图像结构相似比为0.92,表明T-OSEM算法对于动态图像重建是一个比较好的解决方案。     

基于数据融合的ECT图像重建算法

针对固定电极的ECT成像独立测量值较少和因电极位置的影响而导致重建图像失真等问题,提出了基于旋转电极ECT系统模型和数据融合方法来提高重建图像质量。本模型对16电极的ECT模型进行5次旋转,得到的数据采用两种方式进行处理:一是5组数据分别采用线性反投影(LBP)和修正共轭梯度法(MCG)进行重建,再对重建数据进行主分量分析(PCA)的数据融合;二是直接用PCA数据融合再分别进行LBP和MCG图像重建。实验证明:通过增加测量电容数,结合两种数据处理方式可明显提高重构图像质量,降低成像误差。     

基于滤波反投影的脑磁感应迭代重建算法研究

颅脑磁感应断层成像技术(BMIT)是一种非接触、无创的新兴颅脑医学成像技术,图像重建算法是提高重建图像质量的关键。依据BMIT反投影算法和迭代算法,设计出一套基于滤波反投影的脑磁感应迭代重建方法。首先根据滤波反投影重建算法原理,给出初始电导率分布,其次基于电导率变化敏感性加权计算滤波反投影矩阵,最后利用一步牛顿迭代构成滤波反投影迭代重建算法,通过设置理想条件数G来修正Hessian矩阵,改善重建过程的病态程度,并对待重建数据进行标准化位置校正处理。实验结果表明,该算法成像速度快,重建出的图像具有较高分辨率,能够准确反映成像区域内仿真病变的大小及位置信息,且轮廓清晰,为颅脑磁感应断层成像技术应用于临床监护奠定了基础。     

基于两步迭代收缩法和复数小波的压缩传感图像重构

压缩传感系统利用图像稀疏表示的先验知识,能从少量的观测值中重构原始图像.目前压缩传感系统通常利用只有三个方向的正交小波基表示图像,应用迭代收缩法求解对应的优化问题.该方法的缺点是收敛速度慢,并且重构图像有明显的伪吉布斯效应.针对这一缺点,本文提出了结合双树复数小波稀疏图像表示和两步迭代收缩的图像重构算法,在迭代时利用前两个估计值更新当前值.实验结果表明,本文算法的重构图像视觉效果好,收敛速度比传统的重构算法快.     

Split Bregman iteration based image reconstruction algorithm for electrical capacitance tomography

Practical applications of the electrical capacitance tomography (ECT) rely mainly on the effectiveness of reconstruction algorithms. In this paper the solution of the inverse problem with the focus on the ECT imaging is reformulated to be an optimization problem by introducing a new loss function with regularizes encoding multiple features of solution. An iterative scheme that decomposes a complex optimization problem into several simpler sub-problems is developed to solve the proposed loss function, in which the linearization approximation and the acceleration strategy are introduced to improve numerical performances. Numerical experiments validate the effectiveness of the proposed imaging method in tackling the ECT inverse problem.     

基于反投影的MIT动态图像重建方法研究

磁感应断层成像(magnetic induction tomography,MIT)是一种无创、非接触的新型医学成像技术,图像重建算法是实现MIT快速、精确成像的关键.提出一种改进的反投影图像重建算法,首先根据成像区域的磁场分布,由磁力线确定反投影路径,降低了直线反投影用于磁场成像的定位误差;其次根据MIT电磁关系推导,构建了边界检测数据的修正模型,据此对边界相位差数据进行修正处理,进一步提高了重建图像定位准确性;最后分别对成像区域内扰动目标电导率大小变化及位置变化2种情况,构建了序列重建图像,对该图像序列联合分析获取纵向阻抗变化信息,反映了成像体随时间变化的动态信息.实验结果表明该算法具有重建速度快、定位准确的特性,能够准确反映成像区域内部电导率变化,结合序列图像联合分析实现MIT动态成像.     

特征保持的散乱点集光顺算法

提出了一种特征保持的散乱点集光顺算法.首先,搜索点的k最近邻域,计算出点的Delaunay邻域.然后通过考查点及其二阶邻域之间的几何关系,设计出一种带有抑制函数的双边滤波器,用于噪声点集模型的光顺.实验结果显示,本文的光顺算法不但能够有效地去除噪声,抑制体积收缩,并对特征的保持效果也较为理想.     

An improved genetic local search algorithm for defect reconstruction from MFL signals

This paper presents an improved genetic local algorithm by incorporating the simulated-annealing technique into the perturbation process of the genetic local search algorithm and proposes an improved-genetic-local-search-algorithm-based inverse algorithm for two-dimensional defect reconstruction from the magnetic-flux-leakage signals. In the algorithm, a radial-basis-function neural network is utilized as a forward model, and the improved genetic local search algorithm is used to solve the optimization problem in the inverse problem. Experiments are presented to compare the proposed inverse algorithm with both the canonical-genetic-algorithm-based inverse algorithm and the genetic-local-search-algorithm-based inverse algorithm. The results demonstrate that the proposed inverse algorithm is more accurate and robust to the noise.     

用于两相流测量的ECT图像重构技术研究

电容层析成像技术(ECT)具有非侵入、响应速度快、成本低等优点,是用于两相流参数检测非常有发展潜力的技术之一。而图像重构是ECT系统研究的关键技术。该文利用有限元方法对12电极ECT系统进行建模仿真,进行正问题求解,获得了图像重构的样本数据;引入改进的径向基函数神经网络,建立了ECT图像重构算法,并在MATLAB平台上进行了仿真验证。结果表明,改进的径向基神经网络算法在图像重构准确度及速度方面有了明显提高。     

基于改进信赖域的电容层析成像图像重建算法

针对电容层析成像技术中的"软场"效应和病态问题,提出了一种改进信赖域的新电容层析成像算法。在分析电容层析成像基本原理的基础上,推导出了求解ECT反问题的信赖域算法的计算步骤,同时利用BFGS公式对迭代过程中产生的Hesse矩阵进行校正。在此基础上对信赖域算法的收敛性进行了分析和证明,算法满足收敛条件且重建图像误差小。仿真和实验结果表明,和LBP、Landweber和共轭梯度算法相比,对于简单流型该算法兼备成像质量高、边界均匀稳定等优点,为ECT图像重建算法的研究提供了一个新的方法。     

A deep neural networks-based image reconstruction algorithm for a reduced sensor model in large-scale tomography system

Image reconstruction for soft-field tomography is a highly nonlinear and ill-posed inverse problem. Owing to the highly complicated nature of soft-field, the reconstructed images are always poor in quality. One of the factors that affect image quality is the number of sensors in a tomography system. It is commonly assumed that increasing the number of sensors in a tomography system will improve the ill-posed condition in image reconstruction and hence improve image quality. However, as the number of sensors increases, challenges such as more complicated and expensive hardware, slower data acquisition rates, longer image reconstruction times, and larger sensitivity matrices will arise, resulting in a greater ill-posed condition. Since deep learning (DL) is capable of expressing complex nonlinear functions, the majority of research efforts have been directed toward developing a robust DL-based inverse solver for image reconstruction. However, no study has been conducted to solve the inverse problem and improve the quality of the reconstructed image using a reduced sensor model for a large-scale tomography system. This paper proposed an image reconstruction algorithm based on Deep Neural Networks (DNN) to investigate its feasibility in solving the ill-posed inverse problem caused by the reduced sensor model for a large-scale tomography system. The proposed DNN model is based on a supervised, feed-forward, fully connected, backpropagation network. It comprises an input layer, three hidden layers and an output layer. Also, it was trained using large data samples obtained from COMSOL simulation. The relationship between the scattered electromagnetic field measurement and the corresponding true electromagnetic field distribution vector is determined. During the image reconstruction process, the untrained scattered electromagnetic field measurement samples are used as inputs to the trained DNN model, and the model output is an estimate of the electromagnetic field distribution. The results show that the proposed DNN can accurately describe the distribution of electromagnetic field and boundary shape of phantom compared to traditional algorithms (LBP, FBP, Noser and Tikhonov), regardless of the size and number of phantoms within the monitoring area. Hence, the proposed DNN is more robust and has a high degree of generalization.