Speckle interferometry for measurement of continuous deformations

Improvements of a method for measurement of continuous displacements and deformations with digital phase shifting speckle pattern interferometry are presented. The method is based on an algorithm that, with the knowledge of the initial phase, only needs one image at a time to evaluate continuos phase changes due to object deformations. In the improved method, the initial random phase of the speckle pattern is evaluated using a number of phase-shifted images before the deformation under study. This is used for increasing the accuracy of the initial phase estimation and reducing influences from image noise and other measurement disturbances. The phase-shifted speckle patterns are used as references for comparison with the speckle patterns of the deformed object, thereby increasing the reliability and accuracy of the phase estimations of the deformed patterns. The technique can be used for measuring deformations such as transients and other dynamic events, heat expansion as well as other phenomena where it is difficult to accomplish phase shifting during deformation.     

一种提高共聚焦显微镜信噪比算法的研究

基于共焦显微镜的成像特点,建立了Kalman滤波算法的理论模型,把Kalman滤波方法引入到系统中,提出一种基于图像像素的Kalman滤波算法,并实现了实时化的Kalman滤波器。实验结果表明:该算法能够有效地提高共焦显微镜信噪比,但是以牺牲时间为代价,提高系统分辨率的根本方法还是要着重考虑优化成像系统光路和探测电路。     

核密度估计时-空域滤波红外背景抑制方法

提出一种基于核密度估计的时-空域滤波算法,用于红外搜索跟踪系统图像的背景抑制。算法分为空域滤波和时域滤波两部分。在空域滤波中,采用核密度估计算法对背景进行平滑;在时域滤波中,采用核密度估计算法对经过空域滤波后的图像灰度值进行概率计算,判别属于背景残差的灰度值,然后做进一步的滤除。核方法对背景有很好的光滑性且易于计算机实现,实验表明,这种非参方法设计的时-空域滤波算法对背景杂波有非常良好的抑制效果,信噪比也得到明显提高。     

Path-independent digital image correlation with high accuracy,speed and robustness

The initial guess transferring mechanism is widely used in iterative DIC algorithms and leads to path-dependence. Using the known deformation at a processed point to estimate the initial guess at its neighboring points could save considerable computation time, and a cogitatively-selected processing path contributes to the improved robustness. In this work, our experimental study demonstrates that a path-independent DIC method is capable to achieve high accuracy, efficiency and robustness in full-field measurement of deformation, by combining an inverse compositional Gauss–Newton (IC-GN) algorithm for sub-pixel registration with a fast Fourier transform-based cross correlation (FFT-CC) algorithm to estimate the initial guess. In the proposed DIC method, the determination of initial guess accelerated by well developed software library can be a negligible burden of computation. The path-independence also endows the DIC method with the ability to handle the images containing large discontinuity of deformation without manual intervention. Furthermore, the possible performance of the proposed path-independent DIC method on parallel computing device is estimated, which shows the feasibility of the development of real-time DIC with high-accuracy.     

基于改进的空域相关的多聚焦图像融合

提出了一种简单有效的像素级多聚焦图像融合方法。针对正交小波变换缺乏平移不变性而产生视觉失真的缺陷,采用Atrous算法将原图像分解在不同频率域上。Atrous算法先将滤波器h0(n),h1(n)各点间插入适当的零值后再与低频信号做卷积,故又称为"多孔算法"。将具有抑制噪声性能的空阈相关法作为高频子图像的融合规则,选取相关性强边缘特征显著的点作为最终融合子图像的像素点。实验表明,由此融合的图像能完好的保留边缘纹理信息。融合后的图像在客观评价和主管视觉效果上均有提高。     

列图像中遮挡缺损图像补偿算法

针对动态图像的遮挡缺损问题,提出了基于图像块的帧间预测算法,可以检测出序列图像中的遮挡缺损区域;利用帧间灰度相关性修正补偿算法进行遮挡缺损区域补偿,利用修正的像元对缺损区域进行补偿能够克服传统插值补偿算法补偿痕迹明显的缺点.仿真结果表明:该算法补偿处理区域与邻域灰度过渡平滑,消除了补偿痕迹,补偿效果比较理想.     

基于改进开关中值滤波的多孔网栅图像脉冲噪声消除

针对多孔网栅闪光照相图像含有随机脉冲噪声的问题,提出了一种改进的开关中值滤波噪声消除算法。该算法利用像素与邻域窗口统计中值的灰度信息,建立噪声点探测器。通过设置噪声点探测阈值来识别噪声,并用邻域窗口内统计中值代替噪声点取值。经过多次滤波,含随机脉冲噪声的计算机合成网栅图像及实验网栅图像可获得良好的恢复效果。     

MR fingerprinting reconstruction with Kalman filter

Magnetic resonance fingerprinting (MR fingerprinting or MRF) is a newly introduced quantitative magnetic resonance imaging technique, which enables simultaneous multi-parameter mapping in a single acquisition with improved time efficiency. The current MRF reconstruction method is based on dictionary matching, which may be limited by the discrete and finite nature of the dictionary and the computational cost associated with dictionary construction, storage and matching.In this paper, we describe a reconstruction method based on Kalman filter for MRF, which avoids the use of dictionary to obtain continuous MR parameter measurements. With this Kalman filter framework, the Bloch equation of inversion-recovery balanced steady state free-precession (IR-bSSFP) MRF sequence was derived to predict signal evolution, and acquired signal was entered to update the prediction. The algorithm can gradually estimate the accurate MR parameters during the recursive calculation. Single pixel and numeric brain phantom simulation were implemented with Kalman filter and the results were compared with those from dictionary matching reconstruction algorithm to demonstrate the feasibility and assess the performance of Kalman filter algorithm.The results demonstrated that Kalman filter algorithm is applicable for MRF reconstruction, eliminating the need for a pre-define dictionary and obtaining continuous MR parameter in contrast to the dictionary matching algorithm.     

对数似然图像分割的快速主动轮廓跟踪算法

针对跟踪目标尺度变化问题,提出了基于灰度对数似然图像分割的快速主动轮廓跟踪算法。改进的主动轮廓跟踪算法将根据以目标与背景的颜色差异而建立的对数似然图对图像进行阈值分割和数学形态学处理,再将Kalman滤波器结合到主动轮廓跟踪算法进行目标跟踪。改进的主动轮廓跟踪算法对目标分割准确,轮廓特征显著,跟踪效果稳定,算法能很好地适应跟踪目标尺度变化。通过Kalman滤波器对目标位置点的预测减少了主动轮廓跟踪算法收敛的迭代次数,使算法的运算效率提高了33%左右。     

一种改进的球栅阵列封装焊点射线图像阈值分割算法

球栅阵列封装焊点的射线图像具有信噪比差、背景不均匀等特点,故传统的阈值分割方法无法将目标焊点与背景图像很好的分割.本文通过对球栅阵列封装焊点射线图像直方图的分析,利用了自适应维纳滤波对阈值分割前的图像进行了预处理.根据图像的差异来调整该滤波器的参量,对局部差异大的地方进行小的平滑操作,对局部差异小的地方进行大的平滑操作.在最大类间方差法的基础上,对分割后的图像进行了进一步的分析并提出了改进的二次分割方法.改进的方法为并不直接通过OTSU法进行二值化处理来去除背景,而是在阈值分割得到的两个灰度级内通过计算中值和统计最大灰度像素的方法得到了更优化的阈值,使得去除背景后的焊点图像整体更加清晰和均匀.在背景灰度级内寻找了一个合适的灰度级作为处理后的灰度图像新背景,实验证明该方法明显改进了传统最大类间方差法对球栅阵列封装焊点射线图像的阈值分割效果.     

Initial guess by improved population-based intelligent algorithms for large inter-frame deformation measurement using digital image correlation

Digital image correlation (DIC) has received a widespread research and application in experimental mechanics. In DIC, the performance of subpixel registration algorithm (e.g., Newton-Raphson method, quasi-Newton method) relies heavily on the initial guess of deformation. In the case of small inter-frame deformation, the initial guess could be found by simple search scheme, the coarse-fine search for instance. While for large inter-frame deformation, it is difficult for simple search scheme to robustly estimate displacement parameters and deformation parameters simultaneously with low computational cost. In this paper, we proposed three improving strategies, i.e. Q-stage evolutionary strategy (T), parameter control strategy (C) and space expanding strategy (E), and then combined them into three population-based intelligent algorithms (PIAs), i.e. genetic algorithm (GA), differential evolution (DE) and particle swarm optimization (PSO), and finally derived eighteen different algorithms to calculate the initial guess for qN. The eighteen algorithms were compared in three sets of experiments including large rigid body translation, finite uniaxial strain and large rigid body rotation, and the results showed the effectiveness of proposed improving strategies. Among all compared algorithms, DE-TCE is the best which is robust, convenient and efficient for large inter-frame deformation measurement.     

Single-step scanner-based digital image correlation (SB-DIC) method for large deformation mapping in rubber

The incremental digital image correlation (DIC) method has been applied in the past to determine strain in large deformation materials like rubber. This method is, however, prone to cumulative errors since the total displacement is determined by combining the displacements in numerous stages of the deformation. In this work, a method of mapping large strains in rubber using DIC in a single-step without the need for a series of deformation images is proposed. The reference subsets were deformed using deformation factors obtained from the fitted mean stress-axial stretch ratio curve obtained experimentally and the theoretical Poisson function. The deformed reference subsets were then correlated with the deformed image after loading. The recently developed scanner-based digital image correlation (SB-DIC) method was applied on dumbbell rubber specimens to obtain the in-plane displacement fields up to 350% axial strain. Comparison of the mean axial strains determined from the single-step SB-DIC method with those from the incremental SB-DIC method showed an average difference of 4.7%. Two rectangular rubber specimens containing circular and square holes were deformed and analysed using the proposed method. The resultant strain maps from the single-step SB-DIC method were compared with the results of finite element modeling (FEM). The comparison shows that the proposed single-step SB-DIC method can be used to map the strain distribution accurately in large deformation materials like rubber at much shorter time compared to the incremental DIC method.     

结合卡尔曼滤波器噪声分析的车道线检测跟踪算法

使用卡尔曼滤波对视频序列图像中的具体信息进行跟踪的研究目前是跟踪方向的一个热点.但是在处理卡尔曼滤波跟踪过程中的过程噪声和测量噪声,大部分研究普遍采用的是初始赋值.通过不断的调整数,达到较好的跟踪效果.但是这样做不但没有遵循原始数据的规律,同时调整参数是一项耗时的工作.基于这个原因,提出了一种对卡尔曼滤波的过程噪声和测量噪声进行预估计的方法并将其应用到车道线跟踪过程中.通过对一部分离线数据进行处理,可以基本估计出系统的噪声参数.最后采用车道线跟踪算法对论文中的方法进行验证,实验证明,提出的参数估计方法在车道线的跟踪过程中达到很好的效果,同时处理每帧的时间为50ms左右,满足了实时性的要求.     

基于优化估计的深度图像修复与误差补偿方法研究

针对Kinect传感器在获取深度图像时存在深度值随机跳变的不准确性问题,基于最优估计的思想,提出卡尔曼滤波与多帧平均法相结合的图像修复方法。首先利用卡尔曼滤波对多幅深度图像进行修复处理,实现Kinect传感器在采集信息过程中随着时间递推,深度值的跳变逐渐趋于平稳的效果;然后基于多幅图像平均法确定最终的深度图像,解决了Kinect获取深度值存在误差导致的不精确问题。实验结果表明,该算法的均方根误差为38.102 5,平均梯度为0.471 3,信息熵为6.191 8,与单幅图像修复效果相比,得到的深度图像边缘更加清晰。     

具有紧支撑正交非张量积小波的图像融合

提出了基于一种新的小波——具有紧支撑、正交性、伸缩矩阵为[^2 0 ^0 2]的非张量积小波的图像融合方法。首先根据非张量积小波理论,利用Daubechies构造的单变量滤波器构造出基于四通道的不可分的小波滤波器组,用此滤波器组对参加融合的图像进行分解,然后对低频部分采用取均值、高频部分采用系数绝对值取大的融合算法对分解子图进行融合,最后重构。并采用熵、交叉熵、互信息、均方根误差和峰值信噪比等指标对该方法进行了客观评价。对可见光图像与红外图像、远红外图像与近红外图像、遥感图像、多聚焦图像和其它多类图像的融合实验结果证明本方法有较好的融合效果,其融合性能与采用同样融合算法的张量积db2小波的融合方法的融合性能相当。     

Large deformation measurement scheme for 3D digital image correlation method

Difficulties often arise for digital image correlation (DIC) technique when serious de-correlation occurs between the reference image and the deformed image due to large deformation. An updating reference image scheme could be employed to deal with large deformation situation, however that will introduce accumulated errors. A large deformation measurement scheme, combining improved coarse search method and updating reference image scheme, is proposed in this paper. For a series of deformation images, the correlation calculation begins with a seed point and spreads out. An improved coarse search method is developed to calculate the initial correlation parameters for the seed point, which guarantees that the correlation calculation can be carried out successfully even in large deformation situation. Only for extremely large deformation, the reference image is updated. Using this method, not only extremely large deformation can be measured successfully but also the accumulated error could be controlled. A polymer material tensile test and a foam compression test are used to verify the proposed scheme. Experimental results show that up to 450% tensile deformation and 83% compression deformation can be measured successfully.     

一种八谱段滤光片成像系统设计

针对多光谱滤光片阵列成像采样率低,原始(Raw)数据稀疏所导致的重建图像模糊,高频信息丢失等问题,提出了一种新八谱段滤光片阵列分布方案,利用基于邻域梯度延伸方法对光谱Raw图像进行重建.首先基于二叉树生成法,在重复排列的4×4阵列中设计了一种等空间概率比的八谱段滤光片分布方案;然后针对传感器直接获取的稀疏Raw图像,计算各谱段采样点的梯度信息,在保持图像结构特征和纹理信息的基础上,利用邻域采样点的像素值和梯度值对未采样点进行重建,从而获得完整的光谱图像信息;最后,基于已重建的八谱段光谱图像,采用伪逆矩阵法重构各像素位置的31波段光谱值.结果表明,相对于主流图像重建方法,本文算法提高了重建八谱段光谱图像的峰值信噪比、复合峰值信噪比,降低了光谱均方差,更好地保留了图像的纹理和边缘,有效降低了多光谱滤光片阵列成像中的颜色伪影和图像模糊等现象.     

一种实时检测遮挡阴影的算法

监控系统中,观测视场被遮挡,会导致显示的图像中存在遮挡阴影,影响对图像的进一步分析和处理.针对该问题,提出了一种实时检测遮挡阴影的算法.该算法基于图像序列的帧间运动特性对图像进行分区处理;利用图像子区域的灰度均值进行帧间预测,用来判断图像序列中被遮挡图像帧和可能的遮挡阴影区域;利用空域滤波器剔除误检区域,确定阴影区域.该算法实时性好,易于工程实现.     

基于特征匹配的快速鲁棒数字稳像

针对手持移动摄像装置拍摄视频序列相邻帧间存在平移、小角度旋转运动,而且易受噪声、光照变化的影响等问题,提出一种基于优化Oriented FAST and rotated BRIEF(ORB)特征匹配的实时鲁棒电子稳像算法。对相邻帧预处理后用Oriented FAST算子检测特征点,再用Rotated BRIEF描述提取的特征点并采用近邻汉明距离匹配特征点对,然后采用级联滤波去除误匹配点对,最后使用迭代最小二乘法(ILSM)拟合模型参量进行运动补偿实现稳像。图像匹配测试和稳像实验结果表明:基于改进的ORB算法的电子稳像方法补偿每一帧的时间均小于0.1 s,定位精度可达亚像素级,能有效补偿帧间平移旋转运动,而且对噪声和光照变化有较强鲁棒性。经稳像处理后,实拍视频质量明显提高,峰值信噪比（PSNR）平均提高了10 db。     

Image encryption using a synchronous permutation-diffusion technique

In the past decade, the interest on digital images security has been increased among scientists. A synchronous permutation and diffusion technique is designed in order to protect gray-level image content while sending it through internet. To implement the proposed method, two-dimensional plain-image is converted to one dimension. Afterward, in order to reduce the sending process time, permutation and diffusion steps for any pixel are performed in the same time. The permutation step uses chaotic map and deoxyribonucleic acid (DNA) to permute a pixel, while diffusion employs DNA sequence and DNA operator to encrypt the pixel. Experimental results and extensive security analyses have been conducted to demonstrate the feasibility and validity of this proposed image encryption method.