A novel approach for feature extraction from a gamma‑ray energy spectrum based on image descriptor transferring for radionuclide identification

This study proposes a novel feature extraction approach for radionuclide identification to increase the precision of identification of the gamma-ray energy spectrum set.For easier utilization of the information contained in the spectra,the vectors of the gamma-ray energy spectra from Euclidean space,which are fingerprints of the different types of radionuclides,were mapped to matrices in the Banach space.Subsequently,to make the spectra in matrix form easier to apply to image-based deep learning frameworks,the matrices of the gamma-ray energy spectra were mapped to images in the RGB color space.A deep convolutional neural network(DCNN)model was constructed and trained on the ImageNet dataset.The mapped gamma-ray energy spectrum images were applied as inputs to the DCNN model,and the corresponding outputs of the convolution layers and fully connected layers were transferred as descriptors of the images to construct a new classification model for radionuclide identification.The transferred image descriptors consist of global and local features,where the activation vectors of fully connected layers are global features,and activations from convolution layers are local features.A series of comparative experiments between the transferred image descriptors,peak information,features extracted by the histogram of the oriented gradients(HOG),and scale-invariant feature transform(SIFT)using both synthetic and measured data were applied to 11 classical classifiers.The results demonstrate that although the gamma-ray energy spectrum images are completely unfamiliar to the DCNN model and have not been used in the pre-training process,the transferred image descriptors achieved good classification results.The global features have strong semantic information,which achieves an average accuracy of 92.76%and 94.86%on the synthetic dataset and measured dataset,respectively.The results of the statistical comparison of features demonstrate that the proposed approach outperforms the peak-searching-based method,HOG,and SIFT on the synthetic and measured datasets.

Fractional Envelope Analysis for Rolling Element Bearing Weak Fault Feature Extraction

The bearing weak fault feature extraction is crucial to mechanical fault diagnosis and machine condition monitoring. Envelope analysis based on Hilbert transform has been widely used in bearing fault feature extraction. A generalization of the Hilbert transform, the fractional Hilbert transform is defined in the frequency domain, it is based upon the modification of spatial filter with a fractional parameter, and it can be used to construct a new kind of fractional analytic signal. By performing spectrum analysis on the fractional envelope signal, the fractional envelope spectrum can be obtained. When weak faults occur in a bearing, some of the characteristic frequencies will clearly appear in the fractional envelope spectrum. These characteristic frequencies can be used for bearing weak fault feature extraction. The effectiveness of the proposed method is verified through simulation signal and experiment data. © 2017 Chinese Association of Automation.

Feature Extraction for Audio Classification of Gunshots Using the Hartley Transform

In audio classification applications, features extracted from the frequency domain representation of signals are typically focused on the magnitude spectral content, while the phase spectral content is ignored. The conventional Fourier Phase Spectrum is a highly discontinuous function;thus, it is not appropriate for feature extraction for classification applications, where function continuity is required. In this work, the sources of phase spectral discontinuities are detected, categorized and compensated, resulting in a phase spectrum with significantly reduced discontinuities. The Hartley Phase Spectrum, introduced as an alternative to the conventional Fourier Phase Spectrum, encapsulates the phase content of the signal more efficiently compared with its Fourier counterpart because, among its other properties, it does not suffer from the phase ‘wrapping ambiguities’ introduced due to the inverse tangent function employed in the Fourier Phase Spectrum computation. In the proposed feature extraction method, statistical features extracted from the Hartley Phase Spectrum are combined with statistical features extracted from the magnitude related spectrum of the signals. The experimental results show that the classification score is higher in case the magnitude and the phase related features are combined, as compared with the case where only magnitude features are used.

半监督空谱局部判别分析的高光谱影像特征提取

为充分利用高光谱影像中蕴含的空谱特征,提出了一种半监督空谱局部判别分析的高光谱影像特征提取算法(S4LFDA)。鉴于高光谱数据集具有空间一致性,首先将像元进行空间重构,保存高光谱数据的近邻关系;其次引入光谱信息散度重构像元间的相似度;为了充分利用大量无标签样本提高算法性能,采用模糊C均值聚类算法对样本进行聚类分析得到伪标签;然后通过增加规范化项到局部力导引算法(FDA)的类内散度矩阵和类间散度矩阵中,以此保持无标签样本的聚类结构一致性;最后通过局部FDA算法来保持有标签样本类间散度最大化和类内散度最小化并求解最佳投影向量。S4LFDA算法既保持了数据集在光谱域的可分性,又保持了像元在空间区域内的近邻关系,合理利用有标签样本及无标签样本,提高了算法的分类性能。在Pavia University和Indian Pines数据集上进行实验,总体分类精度达到95.60%和94.38%。与其他维数约简算法相比,该算法有效提高了地物分类性能。

增强组合差分乘积形态学滤波的轴承故障特征提取方法

针对滚动轴承故障信号的非线性、非平稳、强噪声特性导致的常规时频域特征提取方法受限问题,提出一种增强组合差分乘积形态学滤波的轴承故障特征提取方法。在分析数学形态学4种基本运算的正、负冲击脉冲提取特性的基础上,运用级联、差分、乘积构造的一种新的组合差分乘积算子(combination difference multiply operator,CDMO)具备了同时提取正、负冲击脉冲的能力,并发挥梯度乘积运算对脉冲提取更敏感的优势,实现故障信息的充分提取。引入故障特征频率比指标优化CDMO结构元素参数,修正待处理信号的几何特征,提取与结构元素相匹配的信号特征信息。在CDMO滤波的基础上,借助三阶累积量切片谱技术能够抑制高斯噪声、突出二次耦合分量的优势,准确提取故障特征频率及其倍频,增强轴承故障特征提取能力并抑制噪声干扰。依托2种不同来源的工程实际信号并与经典故障特征提取方法对比分析,验证了所提方法的有效性。

Fault feature extraction of planet gear in wind turbine gearbox based on spectral kurtosis and time wavelet energy spectrum

Planetary transmission plays a vital role in wind turbine drivetrains, and its fault diagnosis has been an important and challenging issue. Owing to the complicated and coupled vibration source, time-variant vibration transfer path, and heavy background noise masking effect, the vibration signal of planet gear in wind turbine gearboxes exhibits several unique characteristics： Complex frequency components, low signal-to-noise ratio, and weak fault feature. In this sense, the periodic impulsive components induced by a localized defect are hard to extract, and the fault detection of planet gear in wind turbines remains to be a challenging research work. Aiming to extract the fault feature of planet gear effectively, we propose a novel feature extraction method based on spectral kurtosis and time wavelet energy spectrum （SK-TWES） in the paper. Firstly, the spectral kurtosis （SK） and kurtogram of raw vibration signals are computed and exploited to select the optimal filtering parameter for the subsequent band-pass filtering. Then, the band-pass filtering is applied to extrude periodic transient impulses using the optimal frequency band in which the corresponding SK value is maximal. Finally, the time wavelet energy spectrum analysis is performed on the filtered signal, selecting Morlet wavelet as the mother wavelet which possesses a high similarity to the impulsive components. The experimental signals collected from the wind turbine gearbox test rig demonstrate that the proposed method is effective at the feature extraction and fault diagnosis for the planet gear with a localized defect.

Feature extraction of ship radiated-noise by 1(1/2)-spectrum

The properties of 1(1/2)-spectrum are proved and the performances are an- alyzed. By means of the spectrum, the basic frequency component of the harmonic signals can be enhanced. Gaussian color noise and symetrical distribution noise can be canceled. And non-quadratic phase coupling harmonic components in harmonic signal can be reduced. The ship radiated-noise is analyzed and its 7 features are extracted by the spectrum. By means of B-P artificial neural network, three type ships are classified according to extracted features. The classification results about the three type ships A, B and C are 90%, 91.3% and 85.7%. respectively.

Passive sonar identification (Ⅲ): Feature extraction and pattern plates of double-frequency spectrum as well as average power spectrum

This series of papers deal with vessel recognition. The project is conducted by using fuzzy neural networks and basing on the spectra of vessel radiated-noise. Based on the studies of a large amount of ship radiated-noise data, which has been collected from actual ships on the sea, effectively recognizable features are extracted. Such features include line-spectrum features, stationary and nonstationary spectrum features as well as rhythm features. Finally the categorization are tested by unknown samples on the sea, including 33 surface vessels, 8 underwater vessels in 30 operating conditions. Methods for memorization and classilication are also explored in the project. Paper (Ⅲ) is the thirird in the series. It deals with the extraction method of modulation information in double-frequency power spectrum and the establishment of pattern plate of double-frequency spectrum as well as average power spectrum. To extract features from double-frequency spectrum, the tendency of wave is subtracted from the wave of each channel and the modulation of high frequency is compensated. The modulation degree of lines is shown by relative Value and converted to fuzzy value by fuzzy function. The pattern-plate of double-frequency spectrum memorises stable line and its respective modulation strength. The pattern-plate of average power spectrum memorizes the spectra mean of typical samples and the standard variance

Feature extraction of ship-radiated noise using higher-order spectrum

The features of the ship noises are analyzed by using the higher-order spectrum (HOS) after studying their distribution. The results show that the different ship noise has different ranges of the main frequency. The main frequencies of the first class ships are less than 120 Hz, while the second class ships drop in 130 Hz -- 320 Hz. The different relationship between w1 and w2 corresponds to different bispectrum graph. There are the same results in the trispectrum. The feature vector is consist of the wls which correspond to the maximum bispectrum B(wl, wl) and the maximum trispectrum B(wl, w1,wl) respectively, the al, w2 which correspond to the maximum bispectrum B(wl, w2).

参数优化FMD的滚动轴承早期故障诊断

由于滚动轴承早期故障信号特征微弱,特征模态分解(feature mode decomposition, FMD)分解性能受参数滤波器长度L和模态个数n的影响,提出一种参数优化FMD早期故障诊断方法。首先,基于平方包络谱基尼系数(square envelope spectrum gini indix, SESGI)自适应确定FMD的滤波器长度L和模态个数n;其次,采用参数优化的FMD将信号分解为n个模态分量,并根据峭度值最大选择敏感模态分量;最后,对敏感模态分量进行包络分析,判断滚动轴承故障类型。仿真和实验结果表明,该方法可以自适应确定FMD最优参数组合,有效提取故障特征信息。通过与变分模态分解(variational mode decomposition, VMD)对比分析,参数优化FMD提取到的故障特征频率倍频较明显,具有更好的特征提取性能,能够实现滚动轴承故障的精确诊断。

基于FWECS-CYCBD的轴承故障特征提取研究

针对最大二阶循环平稳盲解卷积(maximum second-order cyclostationary blind deconvolution,简称CYCBD)特征提取中循环频率和滤波带宽难确定的问题,引入频率加权能量相关谱(frequency weighted energy correlation spectrum,简称FWECS)来改进CYCBD,实现了低信噪比条件下的滚动轴承故障特征提取。首先,通过FWECS获取周期冲击频率,构造循环频率集;其次,利用最大加权谐波显著性指标设计了一种等步长搜索策略,自适应选取滤波器长度;最后,基于优选的循环频率和滤波带宽进行CYCBD解卷积。轴承仿真和实验数据表明:在循环频率等先验信息未知的情况下,FWECS-CYCBD对故障信号中的微弱冲击特征更敏感;与最小熵解卷积、改进最大相关峭度解卷积和自适应最大二阶循环平稳盲解卷积等方法相比,所提方法在低信噪比条件下能较好地提取轴承故障特征频率信息。

基于加权联合提升包络谱的轴箱轴承故障诊断

为解决列车轴箱轴承微弱故障特征在宽频带上难以提取的问题,基于轴承故障信号的二阶循环平稳特性,提出了一种利用加权联合提升包络谱进行故障诊断的方法.首先,利用谱相干算法将振动信号分解到由频谱频率和循环频率构成的双频域,实现振动信号在全频带内的精细化解调,并基于谱相干的局部特征识别轴承候选故障频率;接着,利用1/3二叉树滤波器将频谱频率分割为不同中心频率和带宽的窄带,在窄带内沿着频谱频率对谱相干的模进行积分,得到窄带提升包络谱;然后,以候选故障频率在窄带提升包络谱中的能量占比为诊断性指标,在每一分解层上构造联合提升包络谱;最后,对不同分解层的联合提升包络谱进行加权平均,得到轴承振动信号的加权联合提升包络谱.轨道车辆轴箱轴承台架试验信号的研究结果表明:所提方法的优势在于能充分整合分布于不同窄带内的轴承故障信息,且不依赖于名义故障周期信息;和现有方法相比,能更有效地揭示轴承故障特征频率及其谐波特征,在提取和识别轴箱轴承微弱故障方面具有一定优势.

融合稠密连接网络与MLP-Mixer的频谱感知方法

随着无线应用的持续激增,干扰环境下的通信变得愈发重要,频谱感知技术在解决无线电用频冲突方面发挥了重要作用。然而实际应用环境复杂,获取到的频谱信号不易被高效提取特征,这降低了频谱信号的实用性。如今人工智能在通信领域应用广泛,对通信技术产生重要影响。为此,从深度学习方法入手,提出一种融合稠密连接网络与MLP-Mixer的频谱感知方法。该模型首先通过Deepinsight网络对频谱信号数据实施处理与转换,使其变换为特征图像,再使用生成式对抗网络合成新的特征图,并在得到特征图像后,采用融合稠密连接网络的混合感知器提取特征,从而感知主用户信道占用情况。经过消融试验对比,所提方法相较于已有模型,较好地提升了频谱感知的检测概率。

基于奇异谱分析与决策树的GIS振动缺陷检测方法研究

针对气体绝缘开关设备(gas insulated switchgear,GIS)在运行过程中出现的异常振动问题,文中提出了基于奇异谱分析与梯度提升决策树的GIS振动缺陷检测方法。首先通过加速度传感器采集GIS在不同状态下运行的振动信号,采用奇异谱分析的方法处理所获信号,并提取信号的主导分量,再通过时域分析和频域分析方法提取主导分量的特征参量,最后基于梯度提升决策树算法对特征进行递归与分类分析,建立了缺陷诊断模型以得到最终的GIS振动缺陷检测结果。该方法应用在实验室GIS,成功识别出了屏蔽罩松动与地脚螺栓松动两种缺陷。文中重点对所述方法及其应用进行了详细的介绍,为电气设备的故障检测研究提供了参考。

基于CM-SVDS-SVMD的滚动轴承故障特征提取方法

针对滚动轴承微弱故障特征信息易受噪声干扰提取困难的问题,提出一种新的滚动轴承故障特征提取方法,即协方差矩阵(covariance matrix,CM)、奇异值差分谱(singular value difference spectrum,SVDS)和奇异值中值分解(singular value median decomposition,SVMD)相结合。首先,考虑到旋转机械的故障特征,对轴承故障信号采用1步长方法构造Hankel矩阵;其次,考虑到信号的协方差矩阵对于信号自相关去噪的优势,进而计算Hankel的协方差矩阵并进行空间重构;再次,采用奇异值差分谱方法对重构后的协方差矩阵信号进行分解处理而实现初步降噪,通过奇异值中值分解方法对其进行分解和筛选处理而完成二次降噪,并根据处理后信号的频谱包络,实现轴承故障特征信息的提取;最后,通过滚动轴承仿真数据分析得出,所提方法能够有效提取出噪声信号的故障特征及其谐波,实现不同轴承故障类型特征的有效提取,为滚动轴承故障复杂信号处理和诊断提供了一种新的方法和途径。

基于铁谱分析的煤矿机电设备摩擦磨损状态分析算法研究

介绍一种基于铁谱分析的煤矿机电设备摩擦磨损状态分析算法。该算法主要通过对设备铁谱图像的预处理,提取设备最大块磨损颗粒的多元特征,利用各方面特征的对比,判断机电设备摩擦磨损状态。通过铁谱分析监测煤矿机电设备摩擦磨损状态,为煤矿机电设备运维管理提供支持,确认铁谱分析的应用价值。

多源荧光光谱数据融合下的淡水浮游植物分类识别方法

淡水浮游植物分类识别过程中,主要采用单一的荧光光谱数据进行特征提取,所得特征信息较为片面,分类识别结果的F1分数偏低。为此,以多源荧光光谱数据融合为前提,提出一种新型淡水浮游植物分类识别方法。采用局部线性嵌入算法对多源荧光光谱数据进行降维处理,再通过小波分解算法提取光谱特征信息。运用独立成分分析算法标记出有效的特征信息,依托于多源荧光光谱数据融合原理结合有效特征得到全面的光谱特征信息。将光谱特征输入可解决多分类问题的支持向量机模型,生成淡水浮游植物分类识别结果。实验结果显示,在噪声比例为40%时,文中设计的分类识别方法的分类识别结果F1分数依旧为0.95,与其他两种方法相比提高了14.74%和18.95%,分类结果更加准确。

基于频谱分析的油田注水柱塞泵故障诊断

针对传统注水柱塞泵故障诊断高度依赖经验,实时处理能力不足,准确率低的问题,提出了一种柱塞泵频域信号在线诊断方法。采集了柱塞泵正常和16种故障状态下的振动信号,采用傅里叶变换进行分析,通过建立故障模型库,在线获取振动信号并进行故障特征提取,通过多次连续时间与模型库的均方根误差进行故障诊断,同时对故障中传感器贡献率进行排序并使传感器降维减少计算量。研究结果表明:采用该方法对轴向柱塞泵故障进行诊断,准确率从66.7%提高到80%以上,计算量降低为原来的26.7%。

高光谱影像奇异谱分析特征提取方法:综述与评价

高光谱遥感影像(hyperspectral imagery,HSI)通常包含几十至数百个连续波段,具有图谱合一、光谱连续的特点,能够实现地物的精细分类,被广泛应用农业、林业、城市以及海洋等领域。HSI特征提取是高光谱应用的前提,也是遥感领域的研究热点和前沿课题之一。近年来,奇异谱分析(singular spectrum analysis,SSA)被应用于HSI领域,在光谱特征和空间特征提取方面取得了较好效果,逐渐成为特征提取的一种有效方法。本文首先分析了HSI特征提取的研究进展和存在的问题;其次对SSA方法进行了系统的梳理,分别介绍了光谱域1D-SSA、空间域2D-SSA和光谱-空间组合域SSA 3类方法的作用、效果及优缺点,并在两个公开的HSI数据集和一个高分五号HSI数据上进行了分类效果验证;最后,对SSA特征提取进行了总结,并讨论了未来的研究方向。

基于时频图与改进E-GraphSAGE的网络流量特征提取方法

由于网络系统的时变性,时域空间网络流量不稳定并且分离难度高,传统时空网络模型对时空序列数据空间结构的刻画和对时空特征的挖掘不充分。针对上述问题,文章提出一种基于时频图与改进E-GraphSAGE的网络流量特征提取方法。首先以bior1.3小波基函数为势变基底,完成原始流量一维时域向时频域空间的映射变换,通过可视化分析去除噪声频段;然后在E-GraphSAGE模型的内部融合Conv LSTM模型,构建融合时空长期依赖特征的三维特征提取方法;最后获得包含局部和全局信息的时空频三维特征的边缘嵌入信息,解决了传统时空特征提取模型存在的整体信息缺失问题。可视化分析和分类实验结果表明,处理后的流量特征具有更高的稳定性和可分离度。同时,将文章所提方法与其他关联度较高的方法进行比较,结果表明文章所提方法在准确率、精确度、召回率及F1-score上均表现较好。