基于残差点的干涉相位图自适应滤波

为了能有效地对InSAR干涉相位图进行滤波,利用残差点矩阵作为滤波控制因子,对两种干涉相位图滤波方法进行了改进。该措施在不破坏致密的干涉条纹的情况下,通过引用干涉相位图的残差矩阵,对不同区域的图像采用不同强度的滤波器来实现自适应的降噪,同时详细描述了滤波方法的实施步骤,并用ERS-1重复轨道数据进行了验证,结果表明,该改进方法可取得良好的滤波效果。     

基于多尺度多方向的InSAR自适应相位滤波

在分析InSAR干涉相位图特点的基础上,提出了一种新的自适应相位滤波方法。该方法采用多尺度多方向的滤波窗口,可根据干涉条纹的方向和密度,自适应地选择窗口进行加权中值滤波处理。采用真实的ERS-1/2雷达数据生成的干涉图进行去噪对比试验与分析,表明所提出的方法降噪效果显著,且能较好地保持干涉图的边缘信息。     

数字全变差向量滤波与相位解缠

分析了"噪声"对合成孔径雷达干涉(InSAR)测量中相位解缠的影响,介绍了SAR影像相干斑噪声的固有性、特殊性以及通常的滤波处理方法.基于连续的各向异性扩散偏微分方程图象处理模型,提出适合于干涉图(mterferogram)和干涉相位条纹图(phase fringe)数字全变差滤波(DTVF)模型.计算分析表明滤波处理极大减少了干涉相位条纹图的残差点,降低设置枝切线的复杂性,在保证解缠质量前提下,提高了大幅图象相位解缠可行性.     

一种新的静态小波域干涉相位图滤波方法

干涉相位图中的噪声会妨碍后续的相位解缠,并降低最终的数字高程模型(Digital elevation model,DEM)的精度.该文提出一种静态小波域的干涉相位图滤波方法.该方法通过分块处理大幅干涉相位图,并且自适应地计算贝叶斯门限以分类处理静态小波系数.用SIR-C/X SAR在意大利Etna火山的干涉数据进行实验,并将本文算法处理结果与Goldstein滤波及改进Goldstein滤波的结果相比较.实验结果表明:本文算法能够有效减少干涉相位图中的残余点,具有更好的降噪性能,而且能够更好地改善干涉条纹边缘的清晰程度.     

改进非局部均值滤波的SAR图像降噪

在非局部均值滤波(NLMF)的基础上,通过预生成相似集与2DPCA(two-dimensional principle component analysis)对NLMF进行改进,提出一种新的SAR(synthetic aperture radar)图像降噪方法。在NLMF算法框架下,针对SAR图像噪声的特点,首先经预处理选择邻近的子图像生成相似集,然后通过2DPCA提取子图像的主要特征,此过程减小了斑点噪声对相似性度量的影响,最后在降维后子图像的基础上进行相似性度量。通过仿真SAR图像和真实SAR图像的降噪实验,将本文方法与经典Lee滤波、Kuan滤波、Gamma-Map滤波和NLMF滤波相比较,结果表明,该方法无论在边缘保持还是一致区域的平滑上,都能取得较好的效果,是一种有效的SAR图像降噪算法。     

一种改进的散斑干涉条纹图滤波算法

在现代数字散斑测量方法中,一般采用相减模式来产生数字散斑条纹图,它是进一步图像处理的信息载体,同时不可避免地附带有大量的乘性噪声。为了保证测量精度,在进一步图像处理前必须对其进行降噪,选择合适的滤波方法显得尤为重要。本文介绍了一种新的去噪方法,该方法是基于噪声的概念提出的,简单有效,容易实现。实验结果亦表明该方法比部分中值滤波对散斑干涉条纹图像更有效。     

雷达干涉条纹图滤波方法研究

在合成孔径雷达干涉测量中,为滤除干涉图像中存在的各种噪声,提高相位解缠结果及生成的数字高程模型的精度.从分析干涉条纹图的特点入手,对干涉条纹图的圆周期中值滤波法进行了研究,提出一种基于高斯噪声估计的自适应圆周期中值滤波方法,给出了算法设计及详细的实验数据并与其它滤波方法进行比较.试验结果表明,该方法可以较好地消除干涉图像中存在的噪声,而且保留更多的图像边缘、细节信息.     

SAR图像斑点噪声的滤波方法研究

针对SAR图像的斑点噪声,分析斑点噪声的模型,给出一种改进的加权中值滤波和增强Lee滤波相结合的去噪方法流程,并用VC++实现这一方法流程。通过实例验证表明该方法流程能有效去除SAR图像中的斑点噪声和保持图像边缘信息。     

基于边缘保持的SAR图像预处理方法

SAR图像空间自适应Gamma MAP滤波去噪后,常残留一些类似脉冲噪声的像素点;为了滤除这些噪声,提高图像质量,采用基于阈值的极值中值滤波;在对SAR图像进行预处理时,将基于阈值的极值中值滤波算法级联到Gamma MAP算法滤波后的SAR图像;测试图像和数据显示,该方法提高了SAR图像质量,保留了SAR图像边缘细节,能很好地检测奇异性目标,易于硬件并行架构实现。     

基于Gabor滤波的引导源目标定位

针对引导源目标定位算法在低信噪比声源信号条件下不能有效定位的问题,在推导了引导源目标定位(GTL,Target localization algorithm using a guide source)算法原理基础上,根据低信噪比条件下GTL算法处理结果的特点,提出了利用Gabor滤波对算法输出条纹图进行降噪处理的方法。利用几组不同信噪比的引导声源和目标声源信号,对基于Gabor滤波降噪处理的GTL算法进行了仿真,并比较分析了其降噪前后引导源目标定位算法的处理结果。仿真结果表明,基于Gabor滤波处理后的GTL算法能准确实现低信噪比条件下的目标声源定位。     

一种复合光栅投影的在线相位测量轮廓术

提出了一种复合光栅投影的在线相位测量轮廓术,以解决在像素匹配和相位计算过程中对条纹频率不同需求的矛盾;在像素匹配过程中,提取高频信息以获得质量较高的调制度图像;在相位计算过程中,通过滤波提取低频分量对被测物体进行三维重构,避开高频分量造成的误差;在滤波过程中,设计了一种组合滤波窗,提升了滤波精度;仿真和实验证明了所提方法的有效性和实用性,使用该方法重构物体的误差约为0.89%。     

INSAR数据处理的若干关键技术探讨

合成孔径雷达干涉测量(INSAR)因具有全天时、全天候和高精度的特点而被广泛应用于地学、海洋、资源探测以及灾害监测等众多领域，但仍有若干理论和技术问题还没有得到彻底解决而限制了该项技术的大规模实际应用。本在简要介绍INSAR原理的基础上，重点讨论数据处理中的关键技术，特别着重分析了影响INSAR测量精度的主要环节，包括当前影像匹配、去平地效应、干涉条纹图滤波以及相位解缠中所存在的问题，并给出部分处理实例供学讨论。     

InSAR干涉纹图噪声抑制算法的比较及改进

干涉纹图中的噪声影响着InSAR的图像质量,使得相位解缠无法进行或者使生成的DEM精度降低。为了获得高质量的InSAR纹图,必须对噪声进行有效抑制,同时保持空间分辨率。文中探讨了InSAR干涉纹图的噪声抑制方法,提出了一种新的干涉纹图的滤波方法。并采用了真实数据验证了方法的有效性。     

干涉合成孔径雷达成像技术研究

用SIR-C/X-SAR的X波段的原始数据,实现了真实数据下星载双航过InSAR的数字高度
地图。首先给出了一种新的解释InSAR三维成像原理的数学描述。该描述对成像关系做了合理近
似,与以前的一些描述相比具有如下优点:①把相位差与地表高度在距离线上直接联系在一起,清
晰地反映出它们之间的关系;②指明了基线对高度测量的有效贡献在于其在斜距垂直方向的投影
这一重要概念;③可用于对测高精度与部分成像参数之间的关系做理论分析;④可用于求取相位差
值到地表高度值之间的转换系数;⑤基本反映了成像处理的实际过程;⑥可解释为何工程上可将相
位差与高度之间近似看成线性关系。然后,给出了一个完整的实现数字高度地图的数据处理过程,
并将整个过程分成9个环节,逐一对各环节的目的和作用做了扼要说明。最后,给出了包括单视
SAR成像、干涉相位条纹、水平相差修正和干涉条纹的降噪滤波在内的处理过程中关键步骤的结
果,和最终的三维地貌成像结果。     

基于面阵CCD的迈克尔逊干涉环计数方法

提出一种基于面阵CCD的迈克尔逊干涉环计数方法。该方法利用面阵CCD采集迈克尔逊干涉环图像,采用图像相减、图像滤波、灰度拉伸、边缘检测、二值化、图像膨胀和细化等方法对采集图像预处理,设计有效区域自动搜索算法确定有效区域中心,提出检测梯度方向角变化的方法对干涉环进行计数。实验结果表明,该方法设计的干涉环计数器精度高,稳定性强,满足了微距离测量的实际应用需求。     

A survey of recent developments in parallel implementations of Gaussian elimination

Gaussian elimination is a canonical linear algebra procedure for solving linear systems of equations. In the last few years, the algorithm has received a lot of attention in an attempt to improve its parallel performance. This article surveys recent developments in parallel implementations of Gaussian elimination for shared memory architecture. Five different flavors are investigated. Three of them are based on different strategies for pivoting: partial pivoting, incremental pivoting, and tournament pivoting. The fourth one replaces pivoting with the Partial Random Butterfly Transformation, and finally, an implementation without pivoting is used as a performance baseline. The technique of iterative refinement is applied to recover numerical accuracy when necessary. All parallel implementations are produced using dynamic, superscalar, runtime scheduling and tile matrix layout. Results on two multisocket multicore systems are presented. Performance and numerical accuracy is analyzed. Copyright © 2014 John Wiley & Sons, Ltd.     

A method of image preprocessing based on nonlinear diffusion and information extraction

Thermally induced stresses play a very important role in controlling the structural reliability of microchip packages. In order to evaluate the magnitude of the warpage of measured objects caused by such stresses, the shadow moiré technique was suggested, where how the moiré fringes are extracted is the key feature. In this work, an improved filtering algorithm based on a nonlinear diffusion equation is developed and a medial axis transformation and pruning algorithm are applied to extract the skeleton of the moiré fringe. In order to remove the forficate fringe and isolated noise fringe, a main fringe extraction algorithm is then used, and this is discussed in detail. According to the moiré fringe information extracted from the shadow moiré pattern, the three-dimensional image which indicates warpage of the measured object’s surface is reconstructed. This demonstrates that the measuring accuracy can be improved up to 2.0 μm by using a grating of 100 lines per inch.     

Robust compensation of time delayed plant by continuous‐time high frequency periodic controller with negligible output ripples

The present work addresses the problem of ensuring robust stability to time delayed plants, compensated with continuous‐time high frequency periodic controller. An efficient design methodology is proposed to synthesize the periodic controller for robust compensation of time delayed linear time‐invariant plants. The periodic controller, by virtue of its loop zero‐placement capability, is shown to achieve superior gain as well as phase/delay margin compensation, especially for non‐minimum phase plants having right half plane poles and zeros in close vicinity to each other. The periodic controller is considered in the observable canonical form which results in realizable bounded control input as well as ensuring insignificant periodic oscillations in the plant output. As a consequence, this paper, furthermore, establishes the fact that the periodic controller designed and synthesized with the proposed methodology can be implemented in real‐time with an assurance of model matching and robust zero‐error tracking. Simulation and experimental results are illustrated to establish the veracity of the claims. The closed‐loop system comprising of time‐delayed linear time‐invariant plant with the periodic controller is analyzed employing the averaging principle and presented here explicitly in a meticulous approach. Copyright © 2015 John Wiley & Sons, Ltd.     

Reorientation planning for a multifingered hand based on an orientation states network using regrasp primitives

This paper concerns the reorientative manipulation planning for a grasped object by a multifingered robotic hand. This manipulation is characterized as the one in which the grasped object is reoriented by the hand without being released, in order to achieve a certain desired final grasping orientation. Two primitives (rotation and pivoting) effectively perform this task by sequentially executing them. A planner which generates a sequence of the two primitives is presented. Two salient features of the planner are that it plans by considering the geometrical shape of the object to be manipulated and that based on a proposed states network it works in two phases. In its first off-line phase, executed once only for the given object, it constructs a structured states network composed of all those orientation states of the object at which at least one of the rotation or pivoting primitives is feasible. This states network serves as road-map for sequence planning. The second on-line phase of the planner can be repeatedly provoked to find the sequenced route on the network between any two given states—corresponding to the initial and goal orientation of the grasped object. A program has been developed for the proposed planner.