核主元分析及其在人脸识别中的应用

传统的基于数据二阶统计矩的特征脸法(Eigenface)或主元分析法(PCA)是一种有效的数据特征提取方法,是基于原始特征的一种线性变换。但是,当原始数据中存在非线性属性时,用主元分析法后留下的显著成分就可能不再反映这种非线性属性。而核主元分析则是基于原始数据的高阶统计量,是一种非线性变换,在图像识别中它可以描述多个像素之间的相关性。该文采用KPCA法提取人脸特征,利用线性支持向量机设计分类器,实验结果表明,基于核主元分析方法的识别正确率明显优于基于主元分析法。     

一种基于PCA的数字音频信息隐秘传输方法

描述了一种基于主成分变换（PCA）的数字音频信息隐密传输方法。该方法以彩色图像为寄主图像,对其进行PCA分解,将经压缩、加密、纠错编码后的数字音频码流,隐藏在对彩色图像质量影响最小的PCA变换数据中,生成藏音图像进而实现音频信息的隐密传输。PC机上进行模拟实验的结果表明,该方法在保证图像质量条件下,有较高的嵌入率、较好的可靠性与隐蔽性。     

On‐line Statistical Process Monitoring and Fault Diagnosis in Batch Baker's Yeast Fermentation

This study involves real‐time monitoring and fault diagnosis in batch baker's yeast fermentation. A specific Real Time Statistical Process Analysis and Control (RT‐SPAC) program was developed to monitor instantaneous reaction conditions. The air flow rate fed to the reactor, temperature, pH, and dissolved oxygen concentration in a laboratory‐size fermenter were monitored and recorded by means of on‐line sensors. Under control of the RT‐SPAC program, 22 batch baker's yeast fermentation operations were carried out. In the first 20 operations, an ordinary process was followed under previously defined nominal operating conditions. Historical data collected from these batches were then used for on‐line Dynamic Principal Component Analysis (DPCA) in the course of the following two batches. The last two batches were implemented such that some deliberate faults (in temperature and pH) were introduced during the operation. The results indicated that the software was capable of capturing the process faults, and furthermore the possible causes of these faults were identified by contribution plots.     

主成份分析—分光光度同时测定铜,钴,锌

利用主成份分析法来处理多组分分析中的吸光度数据,同时测定铜、钴、锌。实验结果表明,该方法比最小二乘法更为有效。     

A multivariate statistical controller for on-line quality improvement

Producing good quality products is an important process control objective. However, achieving this objective can be very difficult in a continuous process, especially when quality measurements are not available on-line or they have long time delays. In this paper, a control approach using multivariate statistical models is presented to achieve this objective. The goal of the control approach is to decrease variations in product quality without real time quality measurements. A PCA model which incorporates time lagged variables is used, and the control objective is expressed in the score space of this PCA model. A controller is designed in the model predictive control (MPC) framework, and it is used to control the equivalent score space representation of the process. The score predictive model for the MPC algorithm is built using partial least squares (PLS). The proposed controller can be developed from and implemented on top of existing PID control systems, and it is demonstrated in two case studies, which involve a binary distillation column and the Tennessee Eastman process.     

Analysis of activated sludge process using multivariate statistical tools—a PCA approach

Twelve original physical variables of an activated sludge wastewater treatment system are considered. These cross-correlated variables are transformed in new ones that are not correlated by the use of PCA (principal component analysis), a powerful tool for analysis, monitoring and diagnostics of wastewater treatment processes. Just three principal components explain most of the system total variability (78% of total variance). Thus, the ability to describe the overall characteristics of the process using only three principal components will make the analysis, monitoring and diagnostic of the system easier.

Three groups of variables characterizing the system are detected. The first group identifies variables that represent micro-organisms and inert particulate matter arising from cellular decay, while the second group refers to substrates and flow rate. The third group is related to the pH. Based on these results, the present paper shows how to enlarge the ways of interpreting the characteristics of activated sludge wastewater treatment system.     

Principal component analysis in fast atom bombardment-mass spectrometry of triacylglycerols in edible oils

Positive-ion fast atom bombardment-mass spectrometry, withm-nitrobenzyl alcohol as matrix and a methanolic sodium iodide (Nal) solution, was used to characterize edible oils by their triacylglycerol composition. Cationized triacylglycerols gave characteristic spectra for olive (57 samples), sunflower (9 samples), soybean (9 samples), corn (12 samples), and peanut (10 samples) oils. Relative abundance of [M + Na]⁺ ions showed relevant differences among the oils. The obtained data were analyzed with the aid of chemometrics by using principal component analysis (PCA) to evaluate the differentiation. Twelve variables were reduced to two principal components, accounting for 82% of the total variance. A two-dimensional plot of 97 PCA scores allowed the separation of the edible oils. The analysis is both fast and reproducible.     

Decision Region Connectivity Analysis: A Method for Analyzing High-Dimensional Classifiers

In this paper we present a method to extract qualitative information from any classification model that uses decision regions to generalize (e.g., feed-forward neural nets, SVMs, etc). The method's complexity is independent of the dimensionality of the input data or model, making it computationally feasible for the analysis of even very high-dimensional models. The qualitative information extracted by the method can be directly used to analyze the classification strategies employed by a model, and also to compare strategies across different model types.     

结合主分量分析及Fisher准则的说话人识别方法研究

本文提出了一种基于主分量分析和Fisher准则的新的Mel频率域特征参数。它是在Mel域频谱的基础上做主分量分析，并且根据Fisher准则，按Fisher比的大小进行特征参量的选择而得到的。它充分的利用了各频带间的相关统计信息，能更紧致有效的区分说话人，这样得到的特征矢量，与传统的按相应特征值进行特征选择的方法相比，在相同维数时具有最大的类别区分度。最后我们实现了一个文本无关的说话人自动识别系统，它的后端采用矢量量化实现聚类分析。在语音库上的实验表明本文的特征矢量在说话人识别上比相同维数的传统特征矢量识别率更高，证实了它紧致、区分度好、冗余信息少的优良性能。     

Application of chemometric methods to synchronous UV fluorescence spectra of petroleum oils

Fluorescence spectroscopy offers the possibility to make qualitative and quantitative analysis of crude oils. Its sensitivity and selectivity permit its use for oils exploration and production. The study describes a novel approach for identification and discrimination of oils by chemometric treatment of synchronous ultraviolet fluorescence (SUVF) spectra which are usually studied by indexes method.Clusters of samples are obtained according to their geographic origin on a 3D plane using SUVF indexes and in the space of principal components for PCA method. Evaluation of oils composition is performed with MCR-ALS algorithm by extracting spectra of families representing aromatic structures.A PLS calibration, built with spectra intensities as input and oils origin as output allows to identify oils chosen for the prediction. Results show that mathematical exploration of full spectral region leads rapidly to classify and predict some oils properties according to their geochemistry.