High accuracy digital image correlation powered by GPU-based parallel computing

A sub-pixel digital image correlation (DIC) method with a path-independent displacement tracking strategy has been implemented on NVIDIA compute unified device architecture (CUDA) for graphics processing unit (GPU) devices. Powered by parallel computing technology, this parallel DIC (paDIC) method, combining an inverse compositional Gauss–Newton (IC-GN) algorithm for sub-pixel registration with a fast Fourier transform-based cross correlation (FFT-CC) algorithm for integer-pixel initial guess estimation, achieves a superior computation efficiency over the DIC method purely running on CPU. In the experiments using simulated and real speckle images, the paDIC reaches a computation speed of 1.66×10⁵ POI/s (points of interest per second) and 1.13×10⁵ POI/s respectively, 57–76 times faster than its sequential counterpart, without the sacrifice of accuracy and precision. To the best of our knowledge, it is the fastest computation speed of a sub-pixel DIC method reported heretofore.     

Accuracy of image registration between MRI and light microscopy in the ex vivo brain

A multistep procedure was developed to register magnetic resonance imaging (MRI) and histological data from the same sample in the light microscopy image space, with the ultimate goal of allowing quantitative comparisons of the two datasets. The fixed brain of an owl monkey was used to develop and test the procedure. In addition to the MRI and histological data, photographic images of the brain tissue block acquired during sectioning were assembled into a blockface volume to provide an intermediate step for the overall registration process. The MR volume was first registered to the blockface volume using a combination of linear and nonlinear registration, and two dimensional (2D) blockface sections were registered to corresponding myelin-stained sections using a combination of linear and nonlinear registration. Before this 2D registration, two major types of tissue distortions were corrected: tissue tearing and independent movement of different parts of the brain, both introduced during histological processing of the sections. The correction procedure utilized a 2D method to close tissue tears and a multiple iterative closest point (ICP) algorithm to reposition separate pieces of tissue in the image. The accuracy of the overall MR to micrograph registration procedure was assessed by measuring the distance between registered landmarks chosen in the MR image space and the corresponding landmarks chosen in the micrograph space. The average error distance of the MR data registered to micrograph data was 0.324±0.277 mm, only 8% larger than the width of the MRI voxel (0.3 mm).     

A variational approach to image registration in dynamic contrast-enhanced MRI of the human kidney

Kidney function can be accessed by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) measurements which yield spatially resolved maps of physiological parameters like perfusion or filtration. The motion of the kidneys during the scan is a dominant limitation of the measurement quality, and image registration is necessary for accurate quantification. We analyzed the feasibility of applying an algorithm, originally developed for multimodal registration, to kidney perfusion time series. The algorithm uses a variational calculation scheme to align the images. In four out of five data sets, kidney motion could be reduced to below the spatial resolution of the images of 1.6 mm while preserving the enhancement pattern of kidney perfusion. Fitting a pharmacokinetic model to the data showed an average reduction of the Akaike fit error of 10% for the registered data, suggesting more stable parameters. We conclude that this image registration algorithm is feasible for correcting kidney motion in renal DCE-MRI.     

Improved matrix inversion in image plane parallel MRI

A new 3D parallel magnetic resonance imaging (MRI) method named Generalized Unaliasing Incorporating Support constraint and sensitivity Encoding (GUISE) is presented. GUISE allows direct image recovery from arbitrary Cartesian k-space trajectories. However, periodic k-space sampling patterns are considered for reconstruction efficiency. Image recovery methods such as 2D SENSE (SENSitivity Encoding) and 2D CAIPIRINHA (Controlled Aliasing In Parallel Imaging Results IN Higher Acceleration) are special instances of GUISE where specific restrictions are placed on the k-space sampling patterns used. It is shown that the sampling pattern has large impacts on the image reconstruction error due to noise. An efficient sampling pattern design method that incorporates prior knowledge of object support and coil sensitivity profile is proposed. It requires no experimental trials and could be used in clinical imaging. Comparison of the proposed sampling pattern design method with 2D SENSE and 2D CAIPIRINHA are made based on both simulation and experiment results. It is seen that this new adaptive sampling pattern design method results in a lower noise level in reconstructions due to better exploitation of the coil sensitivity variation and object support constraint. In addition, elimination of the non-object region from reconstruction potentially allows an acceleration factor higher than the number of receiver coils used.     

Thermal feature extraction of servers in a datacenter using thermal image registration

Thermal cameras provide fine-grained thermal information that enhances monitoring and enables automatic thermal management in large datacenters. Recent approaches employing mobile robots or thermal camera networks can already identify the physical locations of hot spots. Other distribution information used to optimize datacenter management can also be obtained automatically using pattern recognition technology. However, most of the features extracted from thermal images, such as shape and gradient, may be affected by changes in the position and direction of the thermal camera. This paper presents a method for extracting the thermal features of a hot spot or a server in a container datacenter. First, thermal and visual images are registered based on textural characteristics extracted from images acquired in datacenters. Then, the thermal distribution of each server is standardized. The features of a hot spot or server extracted from the standard distribution can reduce the impact of camera position and direction. The results of experiments show that image registration is efficient for aligning the corresponding visual and thermal images in the datacenter, and the standardization procedure reduces the impacts of camera position and direction on hot spot or server features.     

EAST运动斯塔克效应诊断数据处理中GPU并行化加速算法的研究

在EAST装置单道运动斯塔克效应(MSE)诊断系统数据处理中,采用CPU(中央处理器)+GPU(图形处理器)异构化模型,实现了数字谐波分析(DHA)算法的并行化加速计算。由CPU完成数据的加载及简单的数学计算,由GPU实现DHA算法的傅里叶正、逆变换及滤波等并行化计算,与串行算法相比,获得了2000倍以上的加速,可以满足MSE诊断实验期间及时数据处理的要求。     

An improvement of image registration based on phase correlation

This paper is dedicated to studying the extended phase correlation method that applies to the sub-pixel evaluation of translational displacement between images. On the basis of experiments, this paper proposes an improved version of classical Hanning window, with involving parameters optimization in the process. We add a window function to the input images in the space domain and a weighting function to the spectrum in the frequency domain. The statistics and distribution of errors prove the applicability of the proposal. The experimental results of the proposed extended phase correlation algorithm show the registration accuracy of 1/100 pixel.     

Digital image correlation in experimental mechanics and image registration in computer vision: Similarities,differences and complements

Digital image correlation and image registration or matching are among the most widely used techniques in the fields of experimental mechanics and computer vision, respectively. Despite their applications in separate fields, both techniques primarily involve detecting the same physical points in two or more images. In this paper, a brief technical comparison of the two techniques is reviewed, and their similarities and differences as well as complements are presented. It is shown that some concepts from the image registration or matching technique can be applied to the digital image correlation technique to substantially enhance its performance, which can help broaden the applications of digital image correlation in scientific research and engineering practice.     

基于算术-几何均值距离的多模态图像配准

根据图像灰度联合概率分布函数与图像相似程度之间的关系,提出了一种基于算术 几何均值距离的多模态图像配准新测度。与基于信息论的测度不同,新测度不再要求概率分布必须满足连续性的要求。实验结果表明,所提出的新测度比基于信息论的测度具有更强的噪声鲁棒性和计算量更小。     

Study of mutual information multimodality medical image registration based on modified simplex optimization method

Because of a different imaging mechanism and highly complexity of body tissues and structures. Different modality medical images provide non-overlay complementary information. This has very important significance for multimodal medical image registration. Image registration is the first and key part of problem to be solved in the integrations. When the spatial position of two medical images is same, the registration could be achieved. For two CT and PET images, the principal axis method is adopted to achieve the rough registration. The modified simplex algorithm is employed to implement global search using the mutual information as similarity measure. The initial registration parameters are achieved through principal axis Based on the results of test, improved simplex method can adjust reflecting distance. Stepped-up optimization algorithm on the new experimental points through the methods of “reflection”, “enlargement”, “shrinkage” or “global systolic”. A mutual information registration based on modified simplex optimization method is presented in this paper to improve the speed of medical image registration.Results indicate that the proposed registration method prevents the optimizing process from falling into local extremum and improves the convergence speed while keeping the precision. The accurate registration of multimodal image with different resolutions is achieved.     

Phase-constrained parallel MR image reconstruction

A generalized method for phase-constrained parallel MR image reconstruction is presented that combines and extends the concepts of partial-Fourier reconstruction and parallel imaging. It provides a framework for reconstructing images employing either or both techniques and for comparing image quality achieved by varying k-space sampling schemes. The method can be used as a parallel image reconstruction with a partial-Fourier reconstruction built in. It can also be used with trajectories not readily handled by straightforward combinations of partial-Fourier and SENSE-like parallel reconstructions, including variable-density, and non-Cartesian trajectories. The phase constraint specifies a better-conditioned inverse problem compared to unconstrained parallel MR reconstruction alone. This phase-constrained parallel MRI reconstruction offers a one-step alternative to the standard combination of homodyne and SENSE reconstructions with the added benefit of flexibility of sampling trajectory. The theory of the phase-constrained approach is outlined, and its calibration requirements and limitations are discussed. Simulations, phantom experiments, and in vivo experiments are presented.     

Methods for quantitative image quality evaluation of MRI parallel reconstructions: detection and perceptual difference model

Many reconstruction algorithms are being proposed for parallel magnetic resonance imaging (MRI), which uses multiple coils and subsampled k-space data, and a quantitative method for comparison of algorithms is sorely needed. On such images, we compared three methods for quantitative image quality evaluation: human detection, computer detection model and a computer perceptual difference model (PDM). One-quarter sampling and three different reconstruction methods were investigated: a regularization method developed by Ying et al., a simplified regularization method and an iterative method proposed by Pruessmann et al. Images obtained from a full complement of k-space data were also included as reference images. Detection studies were performed using a simulated dark tumor added on MR images of fresh bovine liver. Human detection depended strongly on reconstruction methods used, with the two regularization methods achieving better performance than the iterative method. Images were also evaluated using detection by a channelized Hotelling observer model and by PDM scores. Both predicted the same trends as observed from human detection. We are encouraged that PDM gives trends similar to that for human detection studies. Its ease of use and applicability to a variety of MRI situations make it attractive for evaluating image quality in a variety of MR studies.     

多探测器拼接成像系统实时图像配准

依据已设计完成的基于同心球透镜的四镜头多探测器阵列拼接成像系统,对该系统图像拼接配准过程所采用的特征检测提取、特征向量匹配与筛选、空间变换模型参数估计等算法进行了研究。首先,采用Fast-Hessian检测子提取参考图像和待配准图像的特征点,并生成加速鲁棒特征(SURF)描述向量。接着,采用快速近似最近邻(FANN)逼近搜索算法获得初始的匹配点对,并对匹配点对特征向量的欧式距离进行排序。然后,参照成像系统光学设计参数设定合理的阈值,筛选并保留下较好的匹配点对。最后,提出了一种改进的渐进式抽样一致性(IPROSAC)算法对空间变换矩阵模型进行参数估计,从而得到参考图像与待配准图像的空间几何变换关系。实验结果表明:该算法对图像尺寸、旋转和光照变化都具有一定的不变性,特征匹配时间为0.542 s,配准变换时间0.031 s,配准误差精度小于0.1 pixel,可以满足成像系统关于图像配准实时性和准确性的要求,具有一定的工程应用价值。     

非刚体配准在肝癌手术中的应用

提出了一种在由MR图像引导的肝癌手术中对癌症肿块定位的方法。该方法采用非刚体配准技术,将手术前获得的含有清晰肝癌肿块的CT图像与手术中实时获得的MR图像配准,从而确定MR图像中的癌症肿块的位置。该方法分为三个步骤:将CT和MR图像中的肝脏部位分割出来;对分割后的图像采用仿射变换进行粗配准;采用自由形变模型对图像进行精确配准。配准后肝脏表面的距离误差在1.5mm之内;通过视觉检查,其结果令人十分满意。     

扫频光学相干层析视网膜图像配准去噪算法

多帧叠加平均处理是去除扫频光学相干层析系统散斑噪声、获得较为清晰结构信息的有效方法,但眼睛的震颤、漂移、微眼跳等生理特性和系统光路特性会使图像之间存在错位,导致叠加效果不佳、结构稳定性差,为此本文提出一种基于灰度分布信息和目标几何信息相结合的配准算法。该方法根据图像平均灰度分布提取包含目标信息的感兴趣区域,通过相位相关算法和基于分段拟合的灰度投影算法的双重作用校正图像的平移变换;通过拟合视网膜上边界作为特征点迭代确定最佳旋转参数,并再次重新估计平移参数,实现图像的刚性配准;最后通过轴向扫描一对一映射法以能量函数为约束条件实现图像的非刚性配准。对活体兔眼进行实验,结果表明,本文算法配准后的叠加图像边界清晰,结构信息增强,信噪比和对比度平均有效提高一倍多。本算法适用于强噪声视网膜B-Scans图像的配准,能满足多种类型OCT系统的叠加成像需要,具有较高的鲁棒性和图像配准精度。     

改进的SURF彩色遥感图像配准算法

为了进一步提高彩色遥感图像的配准精度,针对遥感图像配准过程中色彩信息利用率低以及误匹配率高的问题,提出一种改进的SURF(Speeded Up Robust Feature)彩色遥感图像配准算法。该算法首先在对彩色遥感图像进行特征点检测基础上,对特征点描述算子进行改进,以使颜色空间变换后得到的特征点色彩信息添加到原描述算子中,并对特征点描述算子进行归一化处理,以增加算子的独特性和对旋转、尺度、光照的鲁棒性;其次,结合单向匹配和双向匹配的比值,提出了一种最优化阈值选择准则,如果欧氏距离比率小于最优化阈值,完成特征点匹配,得到正确匹配点对,再通过变换矩阵得到配准图像。实验结果表明,在保证实时性的条件下,该算法相比于原SURF算法,准确性和稳定性都有一定提高,具有一定的理论和应用价值。     

图像配准中几种特征点提取方法的分析与实验

研究了图像配准中常用的几种特征点提取方法。基于图像配准对特征提取方法的特殊要求,提出了一组量化的特征提取方法评价因子,能用于合理评价各种方法的性能、优点与缺点,为这些方法的改进提供指导性意见和比较依据。通过大量的实验,利用评价因子比较和分析了上述的几种常用特征提取方法的优缺点、适用环境,并提出改进建议。     

Four-operand parallel optical computing using shadow-casting technique

Optical shadow-casting (OSC) technique has shown excellent potential for optically implementing two-operand parallel logic gates and array logic operations. The 16 logic functions for two binary patterns (variables) are optically realizable in parallel by properly configuring an array of 2×2 light emitting diodes. In this paper, we propose an enhanced OSC technique for implementing four-operand parallel logic gates. The proposed system is capable of performing 2¹⁶ logic functions by simply programming the switching mode of an array of 4×4 light emitting diodes in the input plane. This leads to an efficient and compact realization scheme when compared to the conventional two-operand OSC system.     

一种基于改进频域图像配准技术的超分辨力图像处理方法

为提高光电成像系统的空间分辨力,提出了一种基于改进的频率域图像配准技术的超分辨力图像处理方法。首先利用改进的频域图像配准方法估算出低分辨力图像之间的微位移量,然后采用Papoulis-Gerchberg超分辨力处理方法完成图像复原。利用不同重构方法进行了仿真及实验研究,给出了评价参数。模拟和实际显微热图像的处理结果表明:该算法可使图像质量得到改善,分辨的细节更多,可有效地提高光电成像系统的空间分辨力;处理算法简单,计算量小,可实现快速处理。该算法还可应用于其他不可控光学微扫描成像系统中,具有广泛的应用前景。