不同插值算法的光场重聚焦分析

光场数据重聚焦包括空间域重聚焦和频域重聚焦两种方法，在重聚焦过程中均需要进行插值计算，重聚焦精度与插值精度密切相关，而插值算法的复杂度会影响计算效率。在实际大量图像处理中，在重聚焦效果满足精度要求的前提下选择计算效率最高的方法，其中，插值精度对频域影响远大于空间域。介绍了空间域重聚焦和频域重聚焦的原理，以及传统3种插值方法和sinc函数插值法，分别用不同的插值方法对这两种重聚焦方法进行实验，比较其重聚焦效果并计算其效率。实验结果表明，在实际应用中为了满足计算效率的需求，在满足精度要求情况下，空间域重聚焦采用线性插值法最佳，频域采用采样半径为2像素的sinc插值最佳。对同一幅影像多次重聚焦时，频域方法优于空间域方法。     

Synthetic aperture focusing for defect reconstruction in anisotropic media

Ultrasonic inspection plays an important role in numerous industrial fields. One of the prominent tasks with respect to quantitative nondestructive evaluation is the determination of location, shape, size and orientation of defects. In this respect, the synthetic aperture focusing technique (SAFT) has been successfully applied to isotropic materials over the years. In anisotropic media, however, its application suffers from several phenomena, which are the direction dependence of the ultrasonic velocities, the beam skewing effect and the modified transducer radiation characteristics. In this article, a SAFT imaging algorithm is presented which fully accounts for the nature of wave radiation and propagation within anisotropic materials. For three-dimensional defect reconstruction, the spatial dependence of the ultrasonic group velocities as well as the radiation characteristics of the transducer are exploited--respective algorithms have been implemented for orthotropic material symmetry. Tests have been performed on unidirectional composite material.     

Ultrasonic synthetic aperture focusing using planar pulse-echo transducers

The lateral resolution of digital data from the planar (unfocused) pulse-echo transducers used in conventional ultrasonic inspections can be improved using the synthetic aperture focusing technique (SAFT).

For practical applications it is important to minimize the level of sidelobes (artefacts) introduced by SAFT, without significant loss of resolution. This may be achieved by the inclusion of a suitable aperture-weighting function in the SAFT algorithm, combined with a synthetic aperture size related to the width of the transducer beam-spread.

The properties of the resulting optimized SAFT algorithm are quantified using experimental data from a series of artificial flaws (slots) of different sizes.     

Synthetic aperture imaging using sources with finite aperture: deconvolution of the spatial impulse response

A method for ultrasonic synthetic aperture imaging using finite-sized transducers is introduced that is based on a compact, linear, discrete model of the ultrasonic measurement system developed using matrix formalism. Using this model a time-domain algorithm for deconvolution of the transducer's spatial impulse responses (SIRs) is developed that is based on a minimum mean square error (MMSE) criterion. The algorithm takes the form of a spatiotemporal filter that compensates for the SIRs associated with a finite-sized transducer at every point of the processed image. A major advantage of the proposed method is that it can be used for any transducer, provided that its associated SIRs are known. This is in contrast to the synthetic aperture focusing technique (SAFT), which treats the transducer as a point source. The performance of the method is evaluated with simulations and experiments, performed in water using a linear phased array. The results obtained using the proposed method are compared to those obtained with a classical time-domain SAFT algorithm. For a finite aperture source, it is clearly shown that the resolution obtained using the proposed method is superior to that obtained using the SAFT algorithm.     

Experiences with synthetic aperture focusing technique in the field

The detection and evaluation of defects in industrial components relies strongly on ultrasonic inspection techniques. Distance gain size (DG) or reference reflector methods can be improved concerning their localization, signal-to-noise ratio and sizing accuracy by the synthetic aperture focusing technique (SAFT). To obtain a high quality image, parameters like focal probe versus contact technique probe, achieved resolution or features of SAFT images compared with B-scan images are discussed. The implementation of SAFT in a CAD environment allows us to present stacked 2D reconstructions dynamically. On a cladded testblock with half-penny shaped cracks the advantage of combining CAD with SAFT is shown. A 3D SAFT example finalizes the overview of two decades of experience in applying this technique.     

二维圆周光综合孔径阵的优化排列及其成像特性研究

光综合孔径成像的原理是应用干涉原理在空间频率域中进行采样,并通过傅里叶反变换或其他数值变换方法得到空间分辨力远高于单个孔径成像系统的目标图像。由若干个相同的小孔径在二维圆周上优化排列组成的综合孔径成像系统可以在二维空间频率域中实现较为均匀分布的、具有无冗余度的采样点覆盖,为高质量实时成像提供了一个有效的途径。运用模拟退火算法对由7～16个子孔径组成的二维圆周综合孔径阵列进行优化排列。并依据光学衍射成像原理,从空域和频域两个方面详细分析了二维圆周上优化排列与均匀排列光综合孔径阵的成像特性。对7～16个子孔径组成的光综合孔径的仿真结果表明：无论是在空域还是频域上,子孔径直径增大、孔径数目增多以及综合孔径阵的优化排列都是有利于提高成像质量的。但综合后的子孔径的直径的增大,虽然能获得极高的角分辨力,却并不利于成像质量的提高。     

Study of Spatial Hole Burning in Distributed Phase-Shifted Distributed Feedback Semiconductor Lasers

In this paper,the Huygens-Fresnel principle is restudied by means of wavelet analysis firstly,this builds the basis of the theory for wavelet optics.Then the theory of wavelet analysis is used to analyze the filtering in spatial domain and spatial frequency domain.For example,it gives the space variable processing for the filtering with wavelet in spatial domain,the match filtering for the filtering with wavelet in spatial frequency domain and a single slit and a round aperture for the wave-front filtering.The feasibility for treating the problems in optical field with wavelet transform is discussed,and the theoretical frame of wavelet optics is built.     

Non-linear algorithms based on SAFT ideas for reconstruction of flaws

The main idea and the assumptions of time domain synthetic aperture focusing technique (SAFT) are discussed. It is underlined that the major SAFT assumption is that the flaw can be represented by a number of single non-interacting scalar point scatterers, which act as point sources after illumination with the incident wavefield. As a result some important classes of flaws are incorrectly reconstructed by means of SAFT. Among them are cracks from which reflected signals are similar to those from mirror-shaped flaws. Different non-linear SAFT algorithms proposed earlier by the author are discussed and compared. Among them are SAFT with non-linear normalization (NSAFT), second order SAFT, and the radial tomography technique. The principles of these algorithms are discussed. Then their advantages and disadvantages are analysed. Unfortunately each algorithm cannot be considered universal as a classical SAFT, but for many important cases these algorithms are useful for solving inverse problems in ultrasonics. The ideas of non-linear SAFT algorithms are illustrated by physical experiments.     

Calculation of high frequency ultrasonic signals for shear wave insonification in solid material

The goal of the theoretical part is to simulate an automatic ultrasonic inspection with contact technique shear wave probes, where the high frequency signals are captured and used to perform a reconstruction based on the synthetic aperture focusing method "SAFT". Therefore the ultrasonic probe, the scanning path and the defects are parameters in a CAD model. The scattering behavior of the defect is calculated by the Kirchhoff approximation in its elastodynamic version. The result of the simulation--the high frequency data--and the result of the SAFT-reconstructions are compared with experimental results on a steel test block with side drilled and flat bottom holes. The model is validated by the experiment. One of the applications of the model is to identify multiple reflections.     

基于滤波差分的双阈值弱划痕提取算法

高质量光学元件表面缺陷中存在一些深度较浅或者宽度较窄的划痕,在暗场成像检测中,该类划痕产生的散射光灰度值很低,甚至淹没在背景光中,很难被目视或常规机器视觉识别,造成划痕缺陷的漏检。针对该问题,以既有的疵病检测系统为基础,根据划痕灰度的等级特征,提出双阈值法分类处理划痕缺陷。在低阈值的弱划痕处理中,根据弱划痕和背景的频率特征以及空间对比度特征,设计了频域滤波及背景差分算法。通过空间域以及频率域的滤波处理,排除高频噪声以及高亮度噪声,根据几何特征等提取弱划痕图像中的复杂背景。经差分处理后,提取弱划痕并增强对比度,最后与正常灰度级划痕信息一同通过高阈值(正常阈值)进行后续划痕的特征提取,即得到所有的划痕信息,为划痕缺陷总长度计算以及最大长度的分级判定奠定基础。实验结果表明,该算法避免了过低二值化阈值引入的背景等不规则噪声,使得划痕与背景的对比度大大增强。目前该算法已经应用于惯性约束聚变系统中大口径光学表面划痕的定量检测,并且使长度计量的准确度已提升到约80%。     

浅海非均匀水平阵宽带声场信号无源孔径扩展

大孔径水平阵对于浅海低频水声物理实验研究和应用至关重要,然而受实际情况制约,通常使用的水平基阵孔径十分有限,通过孔径扩展处理来扩展基阵孔径是一条重要途径。传统的无源孔径扩展方法是建立在均匀线阵、匀速相对运动和相干窄带连续信号的基础上的,难以适用于非均匀阵以及非连续宽带声源的情况。针对这些问题,提出了非均匀阵宽带声场信号的无源孔径扩展方法。使用静止布设在海底的非均匀水平短阵接收运动声源重复发射的宽带声信号,先开展均匀空间插值,然后在阵元域和波束域进行宽带扩展孔径处理。仿真和实验结果表明,在水深约70 m的浅海波导环境中,纵向间隔27.5 m的2个阵元接收20~200 Hz宽带声场,其空间插值结果与真值的相关系数大于0.99,说明宽带声场插值方法的合理性。在水平非均匀、纵向孔径250 m的短阵接收声场无法分析简正波频散特征的情况下,仿真和实验数据经过宽带无源孔径扩展处理得到孔径大于1 km的均匀线阵的声场,能高分辨区分各阶简正波,证明所提方法是有效的。      

小波分析在光学信息处理中的应用

用小波变换分析惠更斯-菲涅耳原理,为小波光学理论描述理论基础．运用该理论,对光学系统的空域滤波、空频域滤波现象进行分析,其中小波在空域滤波以空间可变处理为例,小波在空频域滤波以匹配滤波为例,波前滤波则以夫琅和费单缝、圆孔等为例,分析了小波变换在光学领域处理问题的可行性,初步建立了小波光学的理论框架. 关键词：     

Defect imaging with elastic waves in inhomogeneous-anisotropic materials with composite geometries

Imaging of defects in composite structures plays an important role in non-destructive testing (NDT) with elastic waves, i.e., ultrasound. Traditionally the imaging of such defects is performed using the synthetic aperture focusing technique (SAFT) algorithm assuming homogeneous isotropic materials. However, if parts of the structure are inhomogeneous and/or anisotropic, this algorithm fail to produce correct results that are needed in order to asses the lifetime of the part under test. Here we present a modification of this algorithm which enables a correct imaging of defects in inhomogeneous and/or anisotropic composite structures, whence it is termed InASAFT. The InASAFT is based on the exact modelling of the structure in order to account for the true nature of the elastic wave propagation using travel time ray tracing techniques. The algorithm is validated upon several numerical and real life examples yielding satisfactory results for imaging of cracks. The modified algorithm suffers, though, from the same difficulties encountered in the SAFT algorithm, namely “ghost” images and eventual lack of clear focused images. However, these artifacts can be identified using a forward wave propagation analysis of the structure.     

Engineering the point spread function of layered metamaterials

Layered metal-dielectric metamaterials have filtering properties both in the frequency domain and in the spatial frequency domain. Engineering their spatial filtering response is a way of designing structures with specific diffraction properties for such applications as sub-diffraction imaging, supercollimation, or optical signal processing at the nanoscale. In this paper we review the recent progress in this field. We also present a numerical optimization framework for layered metamaterials, based on the use of evolutionary algorithms. A measure of similarity obtained using Hölder’s inequality is adapted to construct the overall criterion function. We analyse the influence of surface roughness on the quality of imaging.     

Fabry-Perot干涉技术及其在疲劳损坏检测中的应用

从激光超声的特点出发,分析目前主要的超声检测技术,重点研究Fabry-Perot干涉仪的原理和工程应用。共焦Fabry-Perot干涉仪具有只对固体表面的振动速度灵敏、有较大的入射孔径、集光能力强等优点,克服了其它干涉检测技术对工作环境的严格要求,广泛应用于工业生产中。为检测碳钢的疲劳破坏,设计了一个激光超声检测系统,分析不同超声频率和加载不同循环压力造成的疲劳损坏情况。     

Spatially uniform sampling in 4-D EPR spectral-spatial imaging

Four-dimensional EPR imaging involves a computationally intensive inversion of the sampled Radon transform. Conventionally, N-dimensional reconstructions have been carried out with N-1 stages of 2-D backprojection to exploit a dimension-dependent reduction in execution time. The huge data size of 4-D EPR imaging demands the use of a 3-stage reconstruction each consisting of 2-D backprojections. This gives three orders of magnitude reduction in computation relative to a single stage 4-D filtered backprojection. The multi-stage reconstruction, however, requires a uniform angular sampling that yields an inefficient distribution of gradient directions. We introduce a solution that involves acquisition of projections uniformly distributed in solid angle and reconstructs in three 2-D stages with the spatial uniform solid angle data set converted to uniform linear angular projections using 2-D interpolation. Images were taken from the two sampling schemes to compare the spatial resolution and the line width resolution. The degradation in the image quality due to the additional interpolation was small, and we achieved approximately 30% reduction in data acquisition time.     

An amplitude correlation and differencing method for the monitoring of flaws in repeat ultrasonic inspections

A new amplitude correlation and differencing (ACD) method for the detection of flaw changes using digital pulse-echo ultrasonic data is proposed. The ACD method is based on signal processing techniques applied to digitized pulse amplitudes from scans of the same transducer recorded during two separate ultrasonic inspections. Changes in flaws with a wide variety of sizes and other characteristics can be detected, provided the change is in a direction perpendicular to the centre line of the ultrasonic beam.Practical experience of this method was gained by analysis of ultrasonic data from two inspections of a weld, separated by a year. The effectiveness of the ACD method for flaw monitoring is compared with that of the 6 dB drop sizing method, before and after resolution improvement by the synthetic aperture focusing technique (SAFT).     

Photoacoustic microscopy by scanning mirror-based synthetic aperture focusing technique

Photoacoustic imaging with a synthetic aperture focusing technique(SAFT) is an effective method to improve the lateral resolution for out-of-focus regions in scanning microscopy systems, which commonly require a decent motorized scanning stage for a lateral scan of a transducer to obtain a cross-sectional image. In this study, we propose and test a photoacoustic imaging system with a scanning mirror-based SAFT(SM-SAFT) for simple and fast data acquisition, without the need for a physical scan of the transducer. Photoacoustic images of hair phantoms acquired by SM-SAFT are demonstrated, serving as a proof-of-concept experiment to show the feasibility and potential of the proposed approach.     

A fast implementation for computing spatial impulse response of ultrasonic transducers

The spatial impulse response(SIR) method is often used as the 'gold standard5 in simulation of transient acoustic wave fields due to its high accuracy in the linear domain.However, a high sampling frequency is often required in order to achieve the high accuracy. As a result, a large amount of data has to be processed. In this paper a fast approach for computing spatial impulse response is proposed to reduce the computation burden. The proposed approach is developed by employing the relationship of SIRs at observed points and SIRs of the projection points on the transducer surface. Two critical parameters used in the proposed approach, the calculation sampling frequency and the interpolation sampling frequency, are then analyzed.Results show that for a 2.25 MHz rectangular transducer with the size of 5 mmxlO mm,a calculation sampling frequency of 1000 MHz and an interpolation sampling frequency of500 MHz can achieve superior performance while improving the computation efficiency 18 times than the direct solving.     

分数傅里叶变换啁啾滤波

针对常规傅里叶变换所不能解决的啁啾噪声滤除问题，利用Wigner分布函数分析分数傅里叶变换的空域和频域特性，提出在分数傅里叶变换域进行啁啾滤波的方法。并将该方法应用到图像处理中，对分数傅里叶变换滤除一维和二维图像的啁啾噪声进行了计算机仿真，获得了满意的效果，结果表明该方法在图像处理中的有效性。