基于 GPU 加速技术的 C 形臂 Cone Beam CT 三维 图像快速重建方法

平板探测器技术的发展使得锥形束计算机断层扫描技术(Cone Beam Computerized Tomography,CBCT)成为一种重要的成像技术,有着十分广泛的应用.基于C形臂的CBCT,除了具有CBCT的技术优势外,还特别适合在影像引导介入手术中应用.然而,如何在满足手术实时性要求的同时获得高分辨率高质量的三维断层图像,仍是个十分具有挑战性的课题.文章提出一种基于GPU加速技术的C形臂CBCT三维图像快速重建方法:在算法层面应用GPU并行加速技术对重建算法进行优化,在系统层面通过设计分布式系统和延迟隐藏机制,大大提升了由二维投影图像重建三维体数据的效率.在保持重建精度的前提下,优化后的GPU加速的FDK算法极大地提升了重建过程的计算效率.延迟隐藏机制进一步提升了系统的运行效率.在使用90帧投影时,系统效率提升了26%,重建延迟加速了2.1倍;当使用120帧投影时,系统效率提升39%,重建延迟加速达到3.3倍.     

锥束CT的惩罚SART重建算法

联合代数迭代方法（SART）对重建图像空间是无约束的,其迭代到一定次数后,图像空间的噪声会不断增加。为了解决这一问题,针对三维锥束CT情况,研究了一种增加惩罚项的联合代数迭代算法,同时研究了该算法与有序子集结合的方法。计算机仿真试验表明：在锥束CT图像重建中,该方法能够在抑制噪声的同时提高重建图像的收敛速度。     

Observability and stabilisability of networked control systems with limited data rates

This paper investigates the observability and stabilisability problem for linear time-invariant systems, where sensors and controllers are geographically separated and connected by a stationary memoryless digital communication channel. The limited information of the plant state is transmitted over such a channel to the controller. We focus on explaining the effect imposed by limited data rates. A new quantisation, coding and control scheme is presented to minimise the required data rate for observability and stabilisability. Different from prior research, our study shows that, the required data rate is determined by the state prediction error. Namely, the smaller prediction error requires the smaller codeword length, which leads to the smaller data rate. It is shown that, there exists a lower bound on the average data rate above which the system is observable and stabilisable. Illustrative examples are given to demonstrate the effectiveness of the proposed quantisation, coding and control scheme.     

CT图像SART重建技术的CUDA并行实现

在计算机断层扫描(CT)图像重建领域,当投影数据不完备或者含有噪声时,相对于滤波反投影(FBP)算法,联合代数重建方法(SART)能重建出质量更高、更符合临床诊断要求的图像。但SART方法非常耗时,而算法的并行实现是解决这一问题的有效途径之一。提出一种基于nVIDIA通用设备计算架构(CUDA)实现的SART并行运算方法。实验结果表明,该方法在不牺牲重建图像质量的基础上,重建时间大为缩减,更有利于临床应用。     

人体运动数据重构方法进展

运动捕捉系统产生的人体运动数据是标记点在运动序列中的位置数据,用于驱动人体模型产生真实感的动画。在对近几年有关人体运动数据重构的文献进行综合和分析的基础上,首先对人体运动数据重构进行了问题描述,并对人体运动数据在重构过程中难以避免的噪声问题和特征点(虚拟空间中的标记点)缺失问题的研究分别进行了总结和分析;然后对人体运动数据获取的光学式原型捕捉系统开发的研究进行了讨论,评述了人体运动数据驱动人体几何模型的相关研究;最后对未来研究提出了一些展望。     

锥束CT感兴趣区域图像重建研究

针对锥束CT感兴趣区域扫描中存在的截断投影数据图像重建问题,提出用基于迭代的代数重建(ART)算法进行重建。锥束ART算法的缺点是计算量大、重建速度慢。为了提高该算法的重建速度,提出了一种基于多核平台的快速并行图像重建方法。首先将三维重建区域等分为上下两块,相应地,探测器平面也分为上下两部分;然后通过双线性插值计算虚拟探测器投影数据;最后通过多线程技术在多核平台上实现了ART算法的并行重建,在保持较高重建精度的同时取得了约两倍的重建加速比。在此基础上,通过仿真实验对3DShepp-Logan模型不同感兴趣区域进行了重建,实验结果表明,ART算法用于感兴趣区域图像重建是可行的。     

螺旋锥束CT图像同时代数重建的机群并行化

在CT（Computed Tomography）图像重建领域,当投影数据含有噪声或者不完备时,与Feldkamp算法相比,同时代数重建方法（Simultaneous Algebraic Reconstruction Technique,SART）能重建出更高质量的三维图像。但三维SART方法非常耗时,为了减少SART的运行时间,利用工作站机群（Cluster of Workstations,COWs）进行并行加速是一种重要的方法。针对螺旋锥束扫描,对基于体数据划分的并行算法进行了改进。并在安装了MPICH 1.2.5的工作站机群上进行了实验。实验结果表明,该方法达到了和串行算法一样的重建效果,并且减少了重建时间。     

Intelligent machinery health prognostics under variable operation conditions with limited and variable-length data

High accuracy health prognostics are significant to machinery intelligent operation and maintenance. Current data-driven prognostics achieve great success that benefits from amply learning samples. In fact, data scarcity challenge widely exists in machinery prognostics and health management, especially for high-end equipment. This study aims to solve this dilemma and proposes a novel meta learning algorithm reconstructed by classic variable-length prediction mode and attention mechanism, namely meta attention recurrent neural network (MARNN). Specifically, we first develop the encoder-decoder with attention mechanism (EDA) cell to perform episodic learning for the subtask-level upgrade. Then multiple subtasks with EDA as prediction models are aggregated to accomplish meta-level upgrade, thus mining the general degradation knowledge from historical datasets. Finally, cross-domain prognostics tasks can be easily realized through fine-tuning tricks, and three rotating machinery run-to-failed experiments are conducted to prove the generalizations of MARNN, which can obtain desired results even when the on-site adaptation data is reduced to one-twentieth.     

不完全投影数据图像重建的ART算法研究

ART算法是一种经典的图像重建方法,适合于不完全投影数据重建。为了提高重建速度,通常对权因子进行简化,其结果是在重建图像中普遍存在椒盐噪声。提出了一种改善重建质量的方法,在每次迭代后对重建图像进行有选择的平滑,将平滑结果作为下一次迭代的初值,其特点是将图像处理和图像重建相结合。仿真实验表明该方法非常有效,不但提高了重建质量,而且克服了利用Box模板进行平滑所造成的图像模糊现象。     

Making costly manufacturing smart with transfer learning under limited data: A case study on composites autoclave processing

The integration of advanced manufacturing processes with ground-breaking Artificial Intelligence methods continue to provide unprecedented opportunities towards modern cyber-physical manufacturing processes, known as smart manufacturing or Industry 4.0. However, the “smartness” level of such approaches closely depends on the degree to which the implemented predictive models can handle uncertainties and production data shifts in the factory over time. In the case of change in a manufacturing process configuration with no sufficient new data, conventional Machine Learning (ML) models often tend to perform poorly. In this article, a transfer learning (TL) framework is proposed to tackle the aforementioned issue in modeling smart manufacturing. Namely, the proposed TL framework is able to adapt to probable shifts in the production process design and deliver accurate predictions without the need to re-train the model. Armed with sequential unfreezing and early stopping methods, the model demonstrated the ability to avoid catastrophic forgetting in the presence of severely limited data. Through the exemplified industry-focused case study on autoclave composite processing, the model yielded a drastic (88%) improvement in the generalization accuracy compared to the conventional learning, while reducing the computational and temporal cost by 56%.     

Sensor data reconstruction using bidirectional recurrent neural network with application to bridge monitoring

Sensors are now commonly employed for monitoring and controlling of engineering systems. Despite significant advances in sensor technologies and their reliability, sensor fault is inevitable. Sensor data reconstruction methods have been studied to recover the missing or faulty sensor data, as well as to enable sensor fault detection and identification. Most existing sensor data reconstruction methods use only the spatial correlations among the sensor data, but they rarely consider the temporal correlations among the data. Use of temporal correlations among the sensor data can potentially improve the accuracy for reconstructing the data. This paper presents a data-driven bidirectional recurrent neural network (BRNN) for sensor data reconstruction, taking into consideration the spatiotemporal correlations among the sensor data. The methodology is demonstrated using the sensor data collected from the Telegraph Road Bridge located along the I-275 Corridor in Michigan. The results show that the BRNN-based method performs better than other current data-driven methods for accurately reconstructing the sensor data.     

数据受限条件下的多模态处理技术综述

随着多媒体技术的发展,可获取的媒体数据在种类和量级上大幅提升。受人类感知方式的启发,多种媒体数据互相融合处理,促进了人工智能在计算机视觉领域的研究发展,在遥感图像解译、生物医学和深度估计等方面有广泛的应用。尽管多模态数据在描述事物特征时具有明显优势,但仍面临着较大的挑战。1)受到不同成像设备和传感器的限制,难以收集到大规模、高质量的多模态数据集;2)多模态数据需要匹配成对用于研究,任一模态的缺失都会造成可用数据的减少;3)图像、视频数据在处理和标注上需要耗费较多的时间和人力成本,这些问题使得目前本领域的技术尚待攻关。本文立足于数据受限条件下的多模态学习方法,根据样本数量、标注信息和样本质量等不同的维度,将计算机视觉领域中的多模态数据受限方法分为小样本学习、缺乏强监督标注信息、主动学习、数据去噪和数据增强5个方向,详细阐述了各类方法的样本特点和模型方法的最新进展。并介绍了数据受限前提下的多模态学习方法使用的数据集及其应用方向(包括人体姿态估计、行人重识别等),对比分析了现有算法的优缺点以及未来的发展方向,对该领域的发展具有积极的意义。     

四元数样条插值的人体运动数据重构

为了得到平滑的人体动画,提出一种基于四元数的样条插值算法,利用提取的关键帧实现人体运动序列的有效重构。为减少重构误差、加快收敛速度,将已知关键帧集合作为初始条件,通过迭代算法求出样条曲线的控制点集合。利用样条曲线控制点计算贝塞尔曲线控制点,构造贝塞尔样条曲线段,将各段贝塞尔样条曲线段组合,构造一条基于四元数的样条曲线。根据德卡斯特里奥（de Casteljau）算法插值重构人体运动。实验结果表明,该算法在保证执行效率的同时,可得到光滑的插值结果,实现满足视觉要求的人体运动重构。     

使用面积积分模型探究锥束CT功能成像的方法

为解决锥形束电子计算机断层扫描（CBCT）在功能成像中不能直接扫描获取体素时间-密度曲线（TDC）的问题,提出锥束CT功能成像的数学模型,得到灌注参数。首先建立模拟投影数据的面积积分模型,利用新西兰大白兔动态对比增强断层（DCE-CT）图像仿真模拟出锥束CT投影数据;利用模拟的投影数据,建立功能模型,使用最优化数值技术求解模型参数,获得各体素的近似TDC,并与实际测量的TDC对比验证模型的正确性;使用去卷积模型,利用最优化参数技术求解灌注参数。使用面积积分模型的模拟TDC与实际值的相似度达到了82.91%,使用此TDC可计算灌注参数并进行伪彩成像。利用CBCT投影数据和功能模型可得到体素近似TDC及组织的灌注参数,实现CBCT功能成像的目的。     

Urban building reconstruction from raw LiDAR point data

We present a method for automatic reconstruction of the volumetric structures of urban buildings, directly from raw LiDAR point clouds. Given the large-scale LiDAR data from a group of urban buildings, we take advantage of the “divide-and-conquer” strategy to decompose the entire point clouds into a number of subsets, each of which corresponds to an individual building. For each urban building, we determine its upward direction and partition the corresponding point data into a series of consecutive blocks, achieved by investigating the distributions of feature points of the building along the upward direction. Next, we propose a novel algorithm, Spectral Residual Clustering (SRC), to extract the primitive elements within the contours of blocks from the sectional point set, which is formed by registering the series of consecutive slicing points. Subsequently, we detect the geometric constraints among primitive elements through individual fitting, and perform constrained fitting over all primitive elements to obtain the accurate contour. On this basis, we execute 3D modeling operations, like extrusion, lofting or sweeping, to generate the 3D models of blocks. The final accurate 3D models are generated by applying the union Boolean operations over the block models. We evaluate our reconstruction method on a variety of raw LiDAR scans to verify its robustness and effectiveness.     

针对标记数据不足的数据流分类器

大部分数据流分类算法解决了数据流无限长度和概念漂移这两个问题。但是,这些算法需要人工专家将全部实例都标记好作为训练集来训练分类器,这在数据流高速到达并需要快速分类的环境中是不现实的,因为标记实例需要时间和成本。此时,如果采用监督学习的方法来训练分类器,由于标记数据稀少将得到一个弱分类器。提出一种基于主动学习的数据流分类算法,该算法通过选择全部实例中的一小部分来人工标记,其中这小部分实例是分类置信度较低的样本,从而可以极大地减少需要人工标记的实例数量。实验结果表明,该算法可以在数据流存在概念漂移情况下,使用较少的标记数据对数据流训练出分类器,并且分类效果良好。     

下视SAR数据3维表面重建

目的 合成孔径雷达(SAR)因成像方法、几何角度等原因使得采集到的数据具有稀疏性及残缺性,如果直接用其进行建模,不能真实地还原物体。针对下视SAR数据的特点,提出一种在建模过程中能够自动修补稀疏及残缺数据的重建方法。方法 首先引入大津法对3维SAR数据进行预处理,然后将2维图像分割方法中的Chan-Vese模型推广应用到下视SAR数据的表面重建中,在初始表面及轮廓指示函数的求取过程中引入距离函数和内积函数。结果 将本文方法与等值面抽取法的重建结果进行比较,本文方法在重建的过程中能够自动修补空洞,重建出的模型表面更加光滑,能更加真实地反映原物体的特征。结论 可以将本文方法推广应用到稀疏及残缺SAR数据的建模中。     

Accurate spatio-temporal reconstruction of missing data in dynamic scenes

In this paper, the accurate method for texture reconstruction with non-desirable moving objects into dynamic scenes is proposed. This task is concerned to editor off-line functions, and the main criteria are the accuracy and visibility of the reconstructed results. The method is based on a spatio-temporal analysis and includes two stages. The first stage uses a feature points tracking to locate the rigid objects accurately under the assumption of their affine motion model. The second stage involves the accurate reconstruction of video sequence based on texture maps of smoothness, structural properties, and isotropy. These parameters are estimated by three separate neural networks of a back propagation. The background reconstruction is realized by a tile method using a single texton, a line, or a field of textons. The proposed technique was tested into reconstructed regions with a frame area up to 8–20%. The experimental results demonstrate more accurate inpainting owing to the improved motion estimations and the modified texture parameters.     

1种基于遗失数据重构的软测量方法

提出1种遗失数据重构思想下的软测量方法:先采用主元分析(PCA)离线建立所有变量(包括难测变量)的主元模型,实际应用时,将实时的难测变量看作遗失数据,通过遗失数据重构方法估计出难测变量,增加了软测量方法的灵活性.更进一步,在重构遗失数据时,使用马氏距离取代欧几里德距离作为指标,更准确地反映了过程变量之间的相关关系,由此...     

Comparison of image reconstruction by using near‐field and far‐field data for an imperfect conductor

Image reconstruction by using near‐field and far‐field data for an imperfectly conducting cylinder is investigated. A conducting cylinder of unknown shape and conductivity scatters the incident wave in free space and the scattered near and far fields are measured. By using measured fields, the imaging problem is reformulated into an optimization problem and solved by the genetic algorithm. Numerical results show that the convergence speed and final reconstructed results by using near‐field data are better than those obtained by using far‐field data. This work provides both comparative and quantitative information. © 2001 John Wiley & Sons, Inc. Int J RF and Microwave CAE 11: 69–73, 2001.