Monitoring heart and respiratory rates at radial artery based on PPG

To develop an apparatus for simultaneous monitoring of heart and respiratory rates using photoplethysmography (PPG) and Fast Fourier Transform algorithm (FFT), the PPG waveform was measured at the radial artery with an original based on diffuse reflectance mode photoplethysmographic device. An FFT algorithm was used to extract heart and respiratory information from the PPG waveform. The results good agree with conventional method. The correlation coefficient of heart rate and respiratory rate obtained using PPG waveform and conventional method is 0.988 and 0.976, respectively. The PPG waveform and FFT algorithm can contribute to quicker and more accurate measurement of human heart rate and respiration rate. A portable watch-like device can be made via this method and be utilized to monitor heart and respiratory rates in hospitals or daily life care.     

调节水床效应的双梯度有源噪声控制自适应算法*

为了使自适应反馈有源噪声控制系统能够兼顾水床效应,通过约束次级信号的能量以调节噪声放大,设计了一种双梯度算法。当次级信号满足约束,算法沿着最小化误差信号能量的梯度方向迭代,反之,则沿着最小化次级信号能量的方向迭代。在有源降噪耳机实例中的对比结果表明,该算法能调节噪声放大并保持较大的降噪带宽,且运算量没有显著增加。     

An improved quantum principal component analysis algorithm based on the quantum singular threshold method

Quantum principal component analysis (qPCA) is a dimensionality reduction algorithm for getting the eigenvectors corresponding to top several eigenvalues of the data matrix and then reconstructing. However, qPCA can only construct the quantum state contains all the eigenvectors and eigenvalues. In this paper, we present an improved quantum principal component analysis (Improved qPCA) algorithm with a fixed threshold. We can reduce the singular value less than the threshold to 0 and obtain a target quantum state which can be used to get an output similar to qPCA after phase estimation. Compared with qPCA, our algorithm has only the target eigenvalues and the probability that we get each target eigenvalue is greater. Furthermore, our algorithm can serve as an additional regularization method and a subroutine for subsequent data processing.     

基于改进ABC的LSSVM氧化还原电位预测研究

针对氧化还原电位对于生物氧化提金预处理过程的控制和优化具有重要作用,提出了一种基于改进的ABC算法优化最小二乘支持向量机的预测方法。该算法是在标准人工蜂群算法的基础上,通过引入欧氏距离,使得在一定邻域内观察蜂采用不同于雇佣蜂的搜索策略。采用改进的ABC算法优化最小二乘支持向量机的参数,取得最优解并赋予最小二乘支持向量机进行预测。以新疆某金矿的生产数据进行仿真研究,结果表明：基于改进的ABC算法优化的最小二乘支持向量机具有较高的预测精度,该方法能使模型取得较好的预测效果。     

半透射高光谱结合流形学习算法同时识别马铃薯内外部缺陷多项指标

针对马铃薯内外部缺陷多项指标难以同时识别的问题,提出了一种半透射高光谱成像技术采用流形学习降维算法与最小二乘支持向量机(LSSVM)相结合的方法,该方法可同时识别马铃薯内外部缺陷的多项指标。试验以315个马铃薯样本为研究对象,分别采集合格、外部缺陷(发芽和绿皮)和内部缺陷(空心)马铃薯样本的半透射高光谱图像,同时为了符合生产实际,将外部缺陷马铃薯的缺陷部位以正对、侧对和背对采集探头的随机放置方式进行高光谱图像采集。提取马铃薯样本高光谱图像的平均光谱(390~1 040 nm)进行光谱预处理,然后分别采用有监督局部线性嵌入(SLLE)、局部线性嵌入(LLE)和等距映射(Isomap)三种流形学习算法对预处理光谱进行降维,并分别建立基于纠错输出编码的最小二乘支持向量机(ECOC-LSSVM)多分类模型。通过分析和比较建模结果,确定SLLE为最优降维算法,SLLE-LSSVM为最优马铃薯内外部缺陷识别模型,该方法对测试集合格、发芽、绿皮和空心马铃薯样本的识别率分别达到96.83%,86.96%,86.96%和95%,混合识别率达到93.02%。试验结果表明：基于半透射高光谱成像技术结合SLLE-LSSVM的定性分析方法能够同时识别马铃薯内外部缺陷的多项指标,为马铃薯内外部缺陷的快速在线无损检测提供了技术参考。     

An adaptive correlation ratemeter: a new method for Doppler fetal heart rate measurements

For use in a cardiotocograph, ultrasound Doppler fetal signals have been thought to provide merely a limited reliability of the fetal heart rate record because of the difficulty of signal processing to obtain a consistent trigger, although the signal itself is easy to obtain with an excellent signal-to-noise ratio.However, the authors have developed a unique signal processing system using a correlation technique with an automatic adaptation algorithm, which solved almost all of the difficulties associated with the old Doppler cardiotocograph and showed that the ultrasound Doppler fetal signal can also provide heart rate records compatible to that obtained by a fetal ECG system.The authors are sure that their method is a most promising one in the coming age of microprocessor-oriented instrumentation.     

基于空间自适应非凸正则项全变差相干斑噪声抑制

全变差正则化方法是相干斑噪声抑制研究的热点.非凸正则项能够更好地保持图像的边缘、纹理细节信息;空间自适应正则化参数可以根据像素点所在的区域,合理地控制噪声抑制程度,从而提高噪声抑制效果.本文结合非凸正则项和空间自适应正则化参数提出了一种新的全变差相干斑噪声抑制模型,并且给出了一种模型求解的数值算法.数值试验结果表明该模型能够获得更好的相干斑噪声抑制效果.     

虚拟传声器在有源抗噪声护听器的应用

考察了虚拟传声器应用于有源抗噪声护听器的稳定性和降噪性能。理论上分析了虚拟传声器算法的稳定性,并通过引入反馈控制器和设计次级通道模型使得算法在不同护听器佩戴方式下都保持稳定,获得高噪声衰减和高稳定性。实验结果证实了虚拟传声器的应用可以在耳道口获得更高的窄带噪声衰减,而且所采取的措施可以保证护听器在不同佩戴状态下都能稳定。      

基于稀疏子空间的类星体光谱异常特征并行提取与分析

类星体是人类所观测到的最遥远天体,对于了解早期宇宙的演化具有重要科学意义。由于类星体距离地球较远,其红移一般较大,导致在光学观测窗口中只有很少的特征（发射线）,且难以识别。类星体光谱的异常特征提取与分析可对未知类星体的识别,提供有效的判别依据。离群检测作为数据挖掘领域的一个主要研究内容,旨在发现那些稀有、特殊数据对象及异常特征,可作为从海量类星体光谱数据中,发现特殊、未知类星体的一种有效途径和手段。Spark作为新一代大数据分布式处理框架,可为海量天体光谱的有效分析和处理,提供一个高效且可靠的并行编程平台。本文充分利用集群系统和Spark编程模型的强大数据处理能力,提出一种基于稀疏子空间的类星体光谱异常特征并行提取与分析方法,其工作由三个模块组成,即类星体光谱特征约减、类星体光谱的稀疏子空间构造和搜索、类星体光谱异常特征提取并行算法设计与分析。类星体光谱特征约减模块,通过属性相关性分析来识别呈现聚类结构的类星体光谱特征线,这些特征线通常会聚集在稠密区域且对类星体光谱异常特征检测毫无意义。光谱特征约减旨在运行异常特征检测算法之前剪枝类星体光谱的冗余特征线,缩小光谱数据检测范围。类星体光谱的稀疏子空间构造和搜索模块,通过设定的稀疏系数阈值来测量类星体光谱的子空间密度,并采用粒子群优化方法作为稀疏子空间的搜索策略,从而快速、高效地获取类星体的异常特征。在第三个模块中,提出了一种MapReduce框架下的类星体光谱异常数据并行检测算法,该算法由并行化数据约减策略、稀疏子空间并行搜索技术两个MapReduce构成,达到适应海量光谱数据的处理目标。最后对检测出的部分类星体异常特征进行了理论分析、测量及人眼证认,充分说明稀疏子空间可为识别特殊、未知类星体候选源,提供有效支持和有力证据。     

对比变分自编码器的近红外光谱测量及其在液态样品检测中的应用

近红外光谱是热门的食品检测方法之一,对于这种高维光谱数据的分析常常需采用数据降维算法提取其中的特征,然而绝大多数算法都只能针对单个数据集进行分析。虽然已有基于对比学习的对比主成分分析成功应用于不同水果表面农残的近红外光谱检测中,但是该方法只能以线性的方式组合原有特征,特征提取效果存在局限性,并且需要调节对比参数来控制背景集影响,需要消耗更大的时间成本。cVAE(contrastive variational autoencoder)是一种基于对比学习和变分自编码器的改进算法,被用于图像去噪和RNA序列分析中,它仍然具备分析多个数据集的特点,同时因为组合了神经网络的概率生成模型而具备了提取非线性隐含特征的能力。将cVAE算法应用于近红外光谱分析,建立了准确的近红外光谱数据降维模型。在实际验证中,使用cVAE算法对购买的不同品牌和批次纯牛奶中掺假三聚氰胺进行检测。结果表明,使用VAE算法只能区分出不同品牌和批次的纯牛奶,而其中是否掺假三聚氰胺这一重要信息无法表现出来;而使用cVAE算法进行数据分析时,由于添加了背景数据集分离了无关变量,能够清晰的将有无掺假三聚氰胺的样本分类。这说明了,cV...     

Analyzing radiation absorption difference of dental substance by using Dual CT

The purpose of this study was to evaluate the changes of noise and computer tomography (CT) number in each dental substance, by using the metal artefact reduction algorithm; we used dual CT for this study. For the study, we produced resin, titanium, gypsum, and wax that are widely used by dentists. In addition, we made nickel to increase the artefact. While making the study materials, we made sure that there is no difficulty when inserting the substances inside phantom. In order to study, we scanned before and after using the metal artefact reduction algorithm. We conducted an average analysis of CT number and noise, before and after using the metal artefact reduction algorithm. As a result, there was no difference in CT number and noise before and after using the metal artefact reduction algorithm. However, when it comes to the noise value in each substance, wax's noise value was the lowest whereas titanium's noise value was the highest, after applying the metal artefact reduction algorithm. In nickel, CT number and noise value from artefact area showed a decreased noise value when applying the metal artefact reduction algorithm. In conclusion, we assumed that we could increase the effectiveness of CT examination by applying dual energy's metal artefact reduction algorithm.     

无监督学习AE和MVO-DBSCAN结合LIF在煤矿突水识别中的应用

快速准确的识别突水类型和突水来源对煤矿安全开采具有重要意义,激光诱导荧光（LIF）技术在检测中具有快速性和灵敏性,将LIF应用于煤矿突水的检测,再结合模式识别算法,可快速识别出突水来源。目前用于识别水样光谱的算法过于依赖预先建立的水样光谱数据库,当突水水源不在该库中时,易引发误识别。无监督学习算法DBSCAN在聚类时不需样本集的标签和类别信息,能降低对未知类别的误识别,因此把DBSCAN算法用于突水的激光诱导荧光光谱识别,并将MVO用于DBSCAN的参数寻优,省去繁琐的人工参数寻优过程。实验中,从谢桥煤矿采水点获取四个水样,利用像素为2 048的USB2000+光谱仪采集水样的荧光光谱,每种水样采集30组光谱数据。首先,利用无监督学习算法自动编码器(AE)对原始光谱数据降维,以减少光谱数据中冗余信息对聚类的影响,设计的AE的结构是介于浅层和深层之间的多层网络模型,可将原始光谱数据降到2维;为使降维模型具有稀疏性,在传统的AE算法中加入一个Dropout层,由实验可知,加入Dropout层后的降维模型具有较快的收敛速度。将多元宇宙优化(MVO)算法用于DBSCAN参数寻优,在参数寻优过程中,DBSCAN对降维后的水样光谱识别率最高为97.5%,此时参数所对应的取值范围为[0.023 66 0.040 65];为验证AE对水样光谱数据降维的有效性,把归一化后的未降维的光谱数据用于DBSCAN聚类识别,DBSCAN对原始水样光谱的识别率最高为95%,比降维后的后水样光谱识别率低了2.5%,结果表明,使用AE降维光谱数据,可提高DBSCAN对不同光谱的识别率。最后,用监督学习算法K最近邻(KNN)识别降维后的水样光谱,将识别结果和无监督学习算法DBSCAN的识别结果对比,其中训练集选用三种水样,测试集使用四种水样;在测试集中,监督学习算法只能准确地识别训练集所包含的水样类别,但把训练集没有的类别全部识别错误,而DBSCAN能准确的识别出训练集中没有的水样光谱。非线性降维算法AE能实现对高维的水样光谱数据降维,把MVO-DBSCAN用于煤矿突水水源的LIF光谱识别,可有效降低因矿井水源光谱数据库建立不完备而引起的误识别。     

基于属性约简的应用服务器优化算法改进

优化参数配置是优化应用服务器性能的重要方面;基于传统参数调节的优化策略耗时耗力缺乏系统性和规律性;利用模块化思想针对目标决策函数对应用服务器参数进行分类,可构建条件属性约简模型;基于属性约简的应用服务器优化算法,可去除对于目标决策函数相对不重要的参数,并获得相对重要的参数,从而达到锁定目标重点调节,快速提高系统性能的目的;现有的约简模型优化算法多基于经典粗糙集理论,在等价关系的基础上构造分类,容易造成大量的信息破坏和流失;文章通过拓展等价关系到一般二元关系,利用广义粗糙集理论改良了基于模块化思想和属性约简模型的应用服务器优化算法,通过定义辨识函数对条件属性进行约简,再结合依赖度计算,得到最终目标参数。     

基于正交投影散度的高光谱遥感波段选择算法

由于高光谱数据的海量高维特征,对其进行降维处理成为高光谱遥感研究的一个重要问题.波段选择算法由于能够有效地保留原始数据的信息,在高光谱数据降维及后续的遥感识别与分类等方面具有明显的优越性.文章提出了一种基于正交投影散度(OPD)的波段选择方法,该方法继承了正交子空间投影(OSP)算法的特点,通过把原始数据投影到特征空间...     

Effects of directional microphone and adaptive multichannel noise reduction algorithm on cochlear implant performance

Although cochlear implant (CI) users have enjoyed good speech recognition in quiet, they still have difficulties understanding speech in noise. We conducted three experiments to determine whether a directional microphone and an adaptive multichannel noise reduction algorithm could enhance CI performance in noise and whether Speech Transmission Index (STI) can be used to predict CI performance in various acoustic and signal processing conditions. In Experiment I, CI users listened to speech in noise processed by 4 hearing aid settings: omni-directional microphone, omni-directional microphone plus noise reduction, directional microphone, and directional microphone plus noise reduction. The directional microphone significantly improved speech recognition in noise. Both directional microphone and noise reduction algorithm improved overall preference. In Experiment II, normal hearing individuals listened to the recorded speech produced by 4- or 8-channel CI simulations. The 8-channel simulation yielded similar speech recognition results as in Experiment I, whereas the 4-channel simulation produced no significant difference among the 4 settings. In Experiment III, we examined the relationship between STIs and speech recognition. The results suggested that STI could predict actual and simulated CI speech intelligibility with acoustic degradation and the directional microphone, but not the noise reduction algorithm. Implications for intelligibility enhancement are discussed.     

低复杂度的MIMO声呐协方差矩阵重构方法*

多输入多输出声纳在对目标进行测向时会产生复杂的运算量,从而降低算法的测向效率。针对这一问题,提出了一种基于降维变换方法的低复杂度协方差矩阵重构方法。该方法能够抑制噪声,提高目标侧向性能。首先利用降维变换方法对接收信号进行波束形成,获得低维度的协方差矩阵,再对矩阵进行Toeplitz处理,抑制矩阵的相干性。所得到的新的协方差矩阵,通过特征分解获得噪声子空间和信号子空间,利用MUSIC方法进行测向。为了进一步降低运算复杂度,利用阵型所满足的旋转不变性,可以采用ESPRIT算法对目标进行波达方向估计。理论分析和实验结果表明,该方法有效降低了运算复杂度,提高了算法的测向性能。在有限快拍数的情况下,与传统测向方法相比,具有运算速度快,目标分辨力强的特点。     

一种光谱分析中的降维方法

在可见/近红外光谱分析中,提取光谱数据中的有用信息是建立稳健准确模型的前提。ISOMAP是一种有效的提取数据本真维的降维方法,但对噪声和邻域参数都比较敏感。提出了一种改进的ISOMAP有监督降维方法,利用光谱数据本身的相关性指导邻域图的构建,降低对噪声和邻域参数的敏感程度,以正确表达数据的邻域结构。采用该方法对两组光谱数据降维并进行PLS建模,结果表明,改进后的算法消弱了邻域大小的影响,提取出的本真维数更小,同时提高了模型精度。     

自适应双树复小波遥感图像复原

由于遥感图像先验知识难以获取,提出了一种自适应的双树复小波迭代收缩复原算法。该算法根据模糊程度和噪声程度估计正则化参数,并利用经验公式计算收缩阈值。在实际应用中,算法能有效解决两步迭代算法使用固定参数的缺点,从而达到提高图像复原质量的目的。实验表明:相对于两步迭代算法,该算法复原图像的峰值信噪比提高0.64~12.23dB,收敛速度提高1.4~16倍;同时,算法在提高图像复原质量、抑制噪声干扰及减少计算时间方面优势明显。     

A novel algorithm for the reduction of undersampling artefacts in magnetic resonance images

An innovative algorithm is presented which is effective in reducing the truncation artefacts occurring in magnetic resonance images due to missing k-space samples. The algorithm works first by filling the incomplete matrix of coefficients with zeroes and then adjusting, by an iterative process, the missing coefficients by performing a reduction of the undersampling artefacts. Then, this set of coefficients is used as a basis for a superresolution algorithm that estimates the missing coefficients by modeling the data as a linear combination of increasing and decreasing exponential functions using Prony's method. In fact, the Prony's method consists of the interpolation of a given data set with a sum of exponential functions: the MRI signals can be well represented as a sum of exponential functions and the missing data can be extrapolated by this representation. The algorithm has been proven to perform better than either a simple algorithm, which detects and then reduces the undersampling artefacts, or an algorithm that models the measured data with approximation functions. The presented algorithm is quite simple and is applicable both to missing rows (phase-frequency acquisitions) and to radial-missing angle (acquisition from projections) undersampling. Experimental results are reported; comparisons, made between the results obtained using the presented algorithm and the alternative methods described above, clearly demonstrate the superiority of the algorithm.