A new wavelet transform for reliability-guided phase unwrapping of optical fringe patterns

A new daughter wavelet definition for reliability-guided phase unwrapping of the optical fringe pattern is introduced. The new daughter wavelet definition is given by normalized mother wavelet with a new factor including a Gaussian function. The modulus of the wavelet transform coefficients, obtained by using daughter wavelet under this new daughter wavelet definition, includes not only modulation information but also local frequency information of the deformed fringe pattern. Therefore, it can be treated as a good parameter which represents the reliability of the reconstructed phase data. Mathematical demonstration of this parameter is given. Both the computer simulation and experiment reveal the validity of the proposed method.     

Secret shared multiple-image encryption based on row scanning compressive ghost imaging and phase retrieval in the Fresnel domain

A multiple-image encryption method is proposed that is based on row scanning compressive ghost imaging, (t, n) threshold secret sharing, and phase retrieval in the Fresnel domain. In the encryption process, after wavelet transform and Arnold transform of the target image, the ciphertext matrix can be first detected using a bucket detector. Based on a (t, n) threshold secret sharing algorithm, the measurement key used in the row scanning compressive ghost imaging can be decomposed and shared into two pairs of sub-keys, which are then reconstructed using two phase-only mask (POM) keys with fixed pixel values, placed in the input plane and transform plane 2 of the phase retrieval scheme, respectively; and the other POM key in the transform plane 1 can be generated and updated by the iterative encoding of each plaintext image. In each iteration, the target image acts as the input amplitude constraint in the input plane. During decryption, each plaintext image possessing all the correct keys can be successfully decrypted by measurement key regeneration, compression algorithm reconstruction, inverse wavelet transformation, and Fresnel transformation. Theoretical analysis and numerical simulations both verify the feasibility of the proposed method.     

Analysis on fringe pattern demodulation by use of 2-D CWT

We discuss the wavelet transform profilometry based on the continuous wavelet transform technique as viewed from frequency analysis. We deduce the expression of one-dimensional (1-D) and two-dimensional (2-D) wavelet transform in frequency domain and analyze their characteristics in the application of demodulating the fringe patterns. We also compare 1-D CWT and 2-D CWT in demodulating the oblique fringe patterns with dual carrier frequency components. When oblique fringe patterns are processed, the direction normal to the grating line and x axis or y axis is not identical. By 1-D CWT, in which wavelet transform is carried out row by row, we cannot obtain the most similarity between local signal and the wavelet functions with different dilation values. While a fan 2-D continuous wavelet transformation can deal with the fringe pattern as a 2-D unit as well as has multi-directions, its advantage is that it can be used to exact the information in the spatial direction. However, its spatial localization ability is not very good, which leads that it is not suitable for demodulating the fringe patterns with high phase variation. Computer simulations and experiments have verified our analysis.     

Comparative analysis of interferogram noise filtration using wavelet transform and spin filtering algorithms

The aim of this paper is to analyze 2D fringe pattern denoising performed by two chosen methods based on quasi-1D two-arm spin filter and 2D discrete wavelet transform (DWT) signal decomposition and thresholding. The ultimate aim of this comparison is to estimate which algorithm is better suited for high-accuracy measurements by phase shifting interferometry (PSI) with the phase step evaluation using the lattice site approach. The spin filtering method proposed by Yu et al. (1994) was designed to minimize possible fringe blur and distortion. The 2D DWT also presents such features due to a lossless nature of the signal wavelet decomposition. To compare both methods, a special 2D histogram introduced by Gutman and Weber (1998) is used to evaluate intensity errors introduced by each of the presented algorithms.     

多类运动想象脑电信号特征提取与分类

针对多类运动想象情况下存在的脑电信号识别正确率比较低的问题,提出了一种将小波包方差,小波包熵和共同空间模式相结合的脑电信号特征提取,输入到支持向量机达到分类目的。首先选择小波包去噪后重要导联的脑电信号,进行小波包分解;然后对通道优化选取的重要导联的每个通道信号计算方差和熵值,对重要导联的每个通道信号的子带系数进行重构后,进行共同空间模式特征提取;最后结合2种不同导联方式所获取的特征向量进行分类。采用BCI2005desc_IIIa中l1b数据,该算法的分类正确率最高达到88.75%,相对2种单一的提取方法分别提高28.27%和6.55%。结果表明该算法能够有效提取特征向量,进而改善多类识别正确率较低的问题。     

Phase retrieval of optical fringe patterns from the ridge of a wavelet transform

A new method for phase retrieval of optical fringe patterns is presented. This method is based on a wavelet transform and is capable of extracting the full 2D phase distribution from a single fringe pattern. An important conclusion that the phase of the optical fringe pattern is equal to the phase of its wavelet transform on the ridge of the wavelet transform is theoretically clarified. The method is compared with the Fourier transform and the integration methods. A numerical simulation and an experimental example of phase retrieval are shown.     

基于小波变换和灰度共生矩阵的轮胎花纹检索

针对在用离散小波变换中提取纹理特征缺少纹理的空间分布特性问题,提出引入方向测度的灰度共生矩阵(Gray Level Co-occurrence Matrix,GLCM)与离散小波分解相互融合的算法,在低频子带上借助方向测度引入权值因子的方法提取灰度共生矩阵的六个统计量,用生成的综合特征来描述轮胎花纹的纹理构成,用欧式距离进行相似性度量。实验结果表明,融合算法能够有效提高检索效率,改进方法的检索效率优于用传统的灰度共生矩阵和小波变换提取纹理方法的检索效率。     

结点阈值小波包变换图像去噪新算法

小波包变换是小波变换的推广,可视为普通小波函数的线性组合,具有灵活的时频分析能力,随着分解层数的增加,小波包分解能够在所有的频率范围聚焦。提出一种应用结点阈值小波包变换的新型图像去噪算法。利用小波包变换对含噪图像进行分解,在图像信号的子带层次上进行结点阈值操作,采用软阈值的方法进行阈值处理,结点噪声采用谱熵法估计,并使用峰值信噪比评估去噪后的图像质量。实验结果表明,相比于使用其它阈值方法的小波包图像去噪算法,该算法具有更好的图像去噪性能。     

基于小波神经网络的光纤陀螺误差补偿方法

为了提高光纤陀螺的测量精度,提出了一种基于小波神经网络的误差补偿方法。首先使用小波分析中的Mallat分解算法提取出陀螺信号中的主趋势项,对其误差余项进行重构。然后将重构信号作为小波神经网络的目标输出,将原始陀螺信号作为训练样本。为了提高小波神经网络的训练速度同时防止其陷入局部极小值,采用增加动量因子和自适应调整学习速率的方法来改进训练方法。训练后建立的神经网络模型对光纤陀螺误差具有良好的估计能力。结果表明,经过小波神经网络方法补偿后,光纤陀螺的输出精度达到了0. 019 4°/s,光纤陀螺的测量性能得到了提高。     

封装西林药瓶残留氧气检测中的谐波基线校正和去噪方法

可调谐激光二极管吸收光谱(TDLAS)技术的成熟和快速非接触气体浓度测量的优点,十分适合用于对封装西林药瓶内残留氧气进行浓度检测。采用TDLAS技术对封装西林药瓶内残留氧气进行浓度检测,检测系统的光路经过空气和玻璃药瓶,玻璃瓶壁对激光的散射和衰减是检测系统的主要干扰,给二次谐波信号的稳定性带来了很大影响。设计和搭建了基于TDLAS的封装西林药瓶残氧量检测系统。针对从系统中提取出的二次谐波信号,提出了一种基于小波变换的基线消除和噪声滤除方法, 解决在残留氧气浓度检测过程中基线漂移和噪声干扰问题,克服玻璃瓶壁对二次谐波信号的干扰,效果明显。选用“sym6”小波,将实验测得的信号进行五层小波分解,根据每一层小波分解得到的低频分量求出相应的基线斜率,对五个基线斜率进行加权平均得到原始信号的基线斜率。由求得的基线斜率,对原始信号经过去基线处理,再进行小波分解和软阈值处理后得到重构信号。对氧气浓度为21%的西林瓶的测量结果表明,处理后谐波信号和理论信号之间的相对误差由处理前的1.26%下降到了0.12%,证明了此方法可以很好地解决在残留氧气浓度检测过程中基线漂移和噪声干扰问题,克服玻璃瓶壁对二次谐波信号的干扰,为氧气浓度测量提供很高质量的信号。     

基于改进经验模态分解域内心动物理特征识别模式分量的心电信号重建

心电图(electrocardiogram,ECG)诊断心脏疾病的严格标准,要求有效地消除噪声并准确地重建ECG信号.经验模式分解(empirical mode decomposition,EMD)方法重建ECG信号中,模式混叠及重建采用模式分量的识别以经验为基础,导致重建ECG信号准确度降低,且方法不具有自适应和通用性.本文首先基于积分均值定理提出一种改进的EMD方法——积分均值模式分解(integral mean mode decomposition,IMMD)方法,经5000个高斯白噪声样本的蒙特卡罗法验证,IMMD方法比EMD具有更优多分辨率分析能力,能够有效地缓解模式混叠.其次,基于ECG信号内固有心动物理特征量识别重建ECG信号所采用的模式分量,具有现实物理意义,因此,方法具有自适应和通用性.经验证,提出方法重建47例ECG信号与原ECG信号的相关系数中:31例优于变分模式分解方法;33例优于Haar小波软阈值法;42例优于集总经验模式分解方法;45例优于EMD方法.相关系数均值为0.8904,方差为0.0071,表现稳定且最优.     

小波包节点分段阈值降噪在水声监听中的应用*

水声目标信号在发送、传播过程中,易受到环境噪声、系统自噪声等影响,因此水声监听过程中目标信号会掺杂大量噪声信息。为提高获取目标信号的准确性和可靠性,降低噪声,在已有小波分析基础上,提出小波包节点相对能量判断最优分解层,最优分解层节点系数分段阈值处理重构方法,实现水声监听信号分频段去噪。将0.1 kHz~8.4 k Hz实验数据按节点频率排序划分为5个强弱不同的频段信号实现消噪提取,结果表明该方法可将噪声信号与目标信号有效分离,与全局单一阈值相比,具有较好降噪能力。该方法打破了小波阈值去噪高频处理的局限性,提高了识别精度,改善了全局单一阈值去噪存在的短板,在鱼类分析识别、舰船监听、深海探测等方面具有一定的推广和应用价值。     

混沌干扰中基于同步挤压小波变换的谐波信号提取方法

针对混沌干扰背景下多个谐波信号的提取问题, 提出了一种基于同步挤压小波变换(SST)的谐波信号抽取方法. 首先利用SST将混沌信号和谐波信号组成的混合信号分解为不同的内蕴模态类函数, 然后利用Hilbert变换对分离出的内蕴模态类函数进行频率识别, 从中分离出各谐波信号. 以Duffing混沌背景为例, 对混沌干扰下多谐波信号的提取进行了实验分析. 实验结果表明: 对于不同频率间隔的多个谐波分量, 本文方法的提取结果都具有较高的精度, 而且所提方法对高斯白噪声的干扰具有较好的鲁棒性, 综合提取效果优于经典的经验模态分解方法.     

Phase-shifting windowed Fourier ridges for determination of phase derivatives

Determination of the phase or phase derivative from interferometric fringe patterns is an important task in optical interferometry. The use of wavelet ridges was recently shown to be an effective method for phase retrieval from a single fringe pattern. One necessary requirement in this method is the need for carrier frequency. In cases when carrier frequency is not available, the novel phase-shifting windowed Fourier ridges method can be used. Phase derivatives with the proper sign can be directly retrieved even in the presence of noise. An application for curvature determination from speckle shearographic fringes demonstrates the effectiveness of the method.     

基于指数尺度间隔连续小波变换的相位提取算法

只需要一幅调制图像的光栅投影测量方法主要有傅里叶变换轮廓术(FTP)、小波变换轮廓术(WTP)等。采用基于指数尺度间隔的连续小波变换与重构方法,提取调制图像的瞬时相位。针对指数尺度间隔连续小波变换,指出了足够大的噪声能够改变小波变换脊的位置,并且该脊向上移动的概率最大。因此,为了重构载频信号,选择脊及其紧邻的较大的那个尺度所对应的小波系数,采用灰度图像阈值分割中最大类间方差法(OTSU),剔除掉幅值较小的系数;针对斑点噪声的影响,对OTSU算法的结果进行了修正;使用修正后的系数集合重构载频信号,并计算该信号的瞬时相位。理论分析和实验结果表明算法有效且具有稳健性。     

小波相位解调轮廓术

针对傅里叶变换轮廓术因混频难以准确提取基频的问题,可将小波变换用于三维形貌相位直接解调。对其原理进行了研究,通过编程实现了小波相位解调。取Morlet复小波作为小波母函数,对调制栅线图逐行进行连续小波变换,从各位置的沿尺度方向的小波变换系数幅值的极值中可直接求取对应的相位数据。由于小波函数具有空域_频率两域的局部化特性,因此它对变形栅具有很强的自适应能力。为验证新方法,对其进行了仿真分析,同时还对石膏半球模型和化妆品瓶进行了实际测量,并分别用傅里叶变换和小波变换进行了处理。结果表明,新方法有效地克服了混频的问题,改善了相位解调效果,提高了测量精度,特别适合于复杂物体的形貌测量。     

变分模态分量降噪后重构的水声弱信号检测

针对能量检测法在低信噪比下对非合作水声探测信号的检测性能显著下降的问题,提出了一种组合变分模态分解和小波变换降噪重构的信号检测方法。以信号分解出的各个本征模态函数的近似熵与互相关系数比值作为分量分类参数,将所得分量分为信号分量、含噪信号分量与噪声分量,然后利用第二代小波变换对含噪信号分量降噪后与信号分量组成重构信号,最后对重构信号进行检测。数值仿真结果表明该方法可以在无先验信息的情况下对CW和LFM信号自适应降噪,信噪比0 dB以下时CW信号重构后信噪比提升约12 dB,宽带LFM信号信噪比提升约8～9 dB,有效提升了低虚警概率下信号的检测概率。湖试结果表明,虚警概率为0.1时检测概率可提升至0.9以上,验证了该方法的有效性。     

Noise reduction for lidar returns using local threshold wavelet analysis

Remote sensing technique of lidar belongs to the category of weak signal extraction under strong background noise. For effectively reducing the noise of lidar return signal, a wavelet analysis method using local threshold value is employed. In the local threshold value wavelet method, different threshold values are used to quantify the high frequency coefficients of every decomposition layer. Both the numerical simulation signal contaminated by random noise of different standard deviation and the practical Mie lidar returns were adopted, and the comparisons among sliding-window method, global threshold method and local threshold method were performed for verifying the feasibility of the local threshold method. Experiment results show that the local threshold wavelet method is a useful de-noising method which shows better effects of noise reduction than other two methods.     

Patch-wise denoising of phase fringe patterns based on matrix enhancement

We propose a new approach for the denoising of a phase fringe pattern recorded in an optical interferometric setup. The phase fringe pattern which is generally corrupted by high amount of speckle noise is first converted into an exponential phase field. This phase field is divided into a number of overlapping patches. Owing to the small size of each patch, the presence of a simple structure of the interference phase is assumed in it. Accordingly, the singular value decomposition (SVD) of the patch allows us to separate the signal and noise components effectively. The patch is reconstructed only with the signal component. In order to further improve the robustness of the proposed method, an enhanced data matrix is generated using the patch and the SVD of this enhanced matrix is computed. The matrix enhancement results in an increased dimension of the noise subspace which thus accommodates more amount of noise component. Reassignment of the filtered pixels of the preceding patch in the current patch improves the noise filtering accuracy. The fringe denoising capability in function of the noise level and the patch size is studied. Simulation and experimental results are provided to demonstrate the practical applicability of the proposed method.     

一种低原子序数元素EDX RF的多次导-样条小波解析方法研究

能量色散X射线荧光（EDXRF）光谱分析待测元素的信息主要反映在能谱的特征峰峰位以及特征峰净峰面积中。对于特征峰的准确检测是EDXRF光谱分析的关键。特征X射线之间的能量在低原子序数元素中相差很小,在实际测量过程中由其他一些因素干扰会导致EDXRF光谱中特征峰产生严重重叠,以EDXRF光谱中低序列元素的重叠峰作为研究对象,提出一种四次导数结合三样条小波变换处理低序列元素重叠峰的新方法。通过数学模型模拟重叠峰检测了该方法的可行性,并仿真了实测X荧光光谱数据进行检测得到良好的效果,通过使用了CIT-3000SY X 荧光元素录井仪实测T铅黄铜数据和混合轻元素数据荧光光谱作为验证。首先,介绍导数法以及三样条小波法分解重叠的原理。导数法阶数越高信号越畸形但可以有效提高重峰分离度,而三样条小波变换对低分离度重峰处理较为无力但能有效的保持峰型。通过Tsallis峰信号模拟重叠峰,模拟出3个峰信号,第1个峰和第2个峰的分离度R=0.33,第2个峰和第3个峰的分离度R=0.67,导数处理后信号任仍具有一部分重叠,但是导数处理后不仅保留了信号的峰位值,且出现了分离度变大的现象,而三样条小波对低分离度重叠峰的分解较为无力,但是对于分离度较大的重叠峰具有较好的效果,信号通过四次导增加分离度再进行三样条小波变换,通过调节样条小波分解层次的次数,然后对分解出的高频信号采取适当的系数进行放大,最后进行信号重构。实验实现了对模拟信号的分解。验证了此方法针对重叠峰分解具有可行性。实验采用分解4层的三样条小波变换以及放大6倍的高频信号。然后,处理仿真K元素的重叠光谱,实现了重叠峰的分解,通过仿真实验表明新方法能准确的识别峰位,结果表明只有1%之内的误差,证明了新方法对X荧光光谱重叠峰分解的适用性。最后用此方法对CIT-3000SY X荧光元素录井仪实测T铅黄铜元素数据以及混合轻元素数据X荧光光谱进行处理,实现了对重叠峰的分解,且分解后的峰位误差控制在1%之内,具有较高的准确率。实验结果证明：四次导数结合三样条小波变换能有效分离重叠峰,并且在处理X荧光光谱的重叠峰分解上具有实用性。