一种自适应的Gabor图像特征抽取和权重选择的人脸识别方法

 为了克服光照、表情变化等因素对人脸识别的影响,本文提出了一种自适应的Gabor图像特征抽取和权重选择的人脸识别方法.该方法首先把每幅人脸图像经过Gabor小波变换后得到的40个不同尺度和方向下的图像都看作是独立的样本,再把不同人脸中的同一尺度和方向的变换结果进行特征重组,得到40个独立地新特征矩阵.为了增强对光照、表情变化的鲁棒性,每一新特征矩阵的识别贡献被本文所提出的自适应权重方法计算得到.其次,对每一新特征矩阵采用离散余弦变化进行降维,并采用了鉴别力量分析方法来选取最有鉴别力的离散余弦变换系数作为特征向量.最后,抽取线性鉴别分析特征进行识别.大量的实验证明了本文所提方法的有效性.     

基于子空间分析的人脸识别方法研究

人脸识别技术是模式识别和机器视觉领域的一个重要研究方向,在众多人脸识别的算法中,基于子空间分析的特征提取方法以其稳定可靠的识别效果成为了人脸识别中特征提取的主流方法之一。本文对目前应用较多的子空间分析方法进行了研究,具体介绍了线性子空间分析方法：主成分分析（PCA）、线性鉴别分析（LDA）、独立主成分分析（ICA）、快速主成分分析（FastICA）等及非线性子空间分析方法：基于核的PCA （KPCA）等的基本思想及其在人脸识别中的研究进展,包括一些新的研究成果。此外,还应用orl及Yale B人脸库对几个基础的子空间方法进行了验证实验。实验结果表明,在几个子空间分析方法中,FastICA算法取得了最高的识别率。最后结合实验结果对各算法的优缺点进行了分析总结。     

A novel virtual samples-based sparse representation method for face recognition

A limited training set usually limits the performance of face recognition in practice. Even sparse representation-based methods which outperform in face recognition cannot avoid such situation. In order to effectively improve recognition accuracy of sparse representation-based methods on a limited training set, a novel virtual samples-based sparse representation (VSSR) method for face recognition is proposed in this paper. In the proposed method, virtual training samples are constructed to enrich the size and diversity of a training set and a sparse representation-based method is used to classify test samples. Extensive experiments on different face databases confirm that VSSR is robust to illumination variations and works better than many representative representation-based face recognition methods.     

Producing virtual face images for single sample face recognition

For single sample face recognition, there are limited training samples, so the traditional face recognition methods are not applicable to this problem. In this paper we propose to combine two methods to produce virtual face images for single sample face recognition. We firstly use a symmetry transform to produce symmetrical face images. We secondly use the linear combination of two samples to generate virtual samples. As a result, we convert the special single sample problem into a non-single sample problem. We then use the 2DPCA method to extract features from the samples and use the nearest neighbor classifier to perform classification. Experimental results show that the proposed method can effectively improve the recognition rate of single sample face recognition.     

Fingerprint image enhancement using data driven Directional Filter Bank

A new method based upon data driven tool, principal component analysis (PCA), for fingerprint enhancement is proposed in this paper. PCA is a very useful statistical technique that has found application in many different fields like image compression, face recognition and is commonly used for finding patterns in data of high dimension. In the proposed method, the input image is first decomposed into directional images using decimation free Directional Filter Bank (DDFB). Then these directional images are normalized. A data driven technique PCA is applied to these normalized directional fingerprint images, which gives the PCA filtered images. These are basically directional images. Then these directional images are reconstructed into one image which is the enhanced one. Simulation results are included illustrating the capability of the proposed method.     

利用近红外及中红外融合技术对小麦产地和烘干程度的同时鉴别

小麦是制作馒头的主要原料之一,小麦中水、蛋白质、淀粉会因产地以及烘干程度的差异而不同,进而影响到加工成馒头的品质。所以实现对小麦产地和烘干程度的快速鉴别就显得尤为重要。感官评定是鉴别小麦产地和烘干程度常用的方法,对比感官评定,光谱分析可以识别样品中的分子结构等信息。基于此,尝试利用近红外和中红外光谱融合技术实现对不同产地和不同烘干程度的小麦同时鉴别。首先选取了两个不同产地的小麦,再利用微波干燥法对两个不同产地的小麦做烘干预处理,使烘干的小麦水含量为12%±0.5%,原麦水含量为18%±0.5%。分别标记为原麦A,烘干A,原麦B,烘干B,再将小麦研磨成粉末,过100目筛网筛选后,置于自封袋中备用。随后分别采集四种小麦样品的近红外和中红外光谱信息,在Matlab 7.10的环境下使用标准正态变量变换(standard normal variable transformation, SNVT)对采集到的原始光谱数据进行预处理,利用主成分分析对预处理后的数据进行降维处理,再结合线性判别分析（linear discriminant analysis,LDA）和支持向量机（support vector machine, SVM）分别建立小麦近红外、中红外光谱数据识别模型。另外利用联合区间偏最小二乘法（synergy interval partial least square, SiPLS）筛选出利用标准正态变量变换(SNVT)预处理后的小麦近红外和中红外光谱数据特征光谱区间,将筛选出的近红外和中红外光谱数据特征光谱区间融合后再结合线性判别分析（LDA）和支持向量机（SVM）建立小麦融合光谱信息的识别模型。然后比较同种光谱数据下利用线性判别分析（LDA）和支持向量机（SVM）建立的小麦识别模型识别率、比较同种建模方法下近红外和中红外光谱数据建立小麦识别模型识别率、比较同种建模方法下光谱数据融合和单一光谱数据建立小麦识别模型识别率。结果表明,同种光谱分析方法,利用SVM建立的四种小麦识别模型识别率高于利用LDA建立的小麦识别模型识别率。同种建模方法,近红外光谱数据建立的小麦识别模型识别率优于中红外光谱数据建立的小麦识别模型识别率。而在同种建模方法下,利用SiPLS筛选出近红外和中红外光谱数据的特征光谱区间数据融合后建立小麦识别模型识别率最高,光谱数据融合后结合LDA建立的小麦识别模型校正集识别率为98.75%,预测集识别率为97.50%;而将此选择的变量结合SVM建立的小麦识别模型的校正集和预测集识别率都达到100.0%。对比利用单一光谱数据建立的小麦识别模型识别率,光谱数据融合之后建立的小麦识别模型识别率得到显著提高,该研究从纵向和横向上全面地比较了光谱数据建立的小麦模型识别率,结果可为更准确地运用光谱融合技术建立小麦产地以及烘干程度识别模型提供参考。     

基于近红外光谱和模式识别技术鉴别大米产地的研究

利用近红外光谱和模式识别技术建立了大米产地的快速鉴别方法。首先对119个地理标志产品响水大米和90个其他产地的大米(即非响水大米)的近红外光谱进行一阶导数和平滑处理,利用主成分分析法(PCA)对数据进行降维,通过前三个主成分的载荷图确定了相关性最大的特征波段(7 700～6 700 cm-1与5 700～4 300 cm-1)。在全波段内,凝聚层次聚类和Fisher’s判别鉴别方法都可以100%正确的鉴别响水大米和非响水大米;对于非响水地区的大米的具体产地判别,聚类分析正确率为91.9%,Fisher’s判别分析方法的正确率为96.7%。同时,在特征波段内,对大米产地聚类分析的准确度高于全波段范围内分析结果,说明选取的特征波段具有较强的代表性,是优化模型的有效方法之一。     

基于DPCA与LSSVM的飞机发动机异常状态识别

针对飞机发动机异常状态识别精度差、效率低和易误诊漏诊等问题,提出了一种基于动态主元分析 (Dynamic Principal Component Analysis, DPCA)和最小二乘支持向量机(Least Square Support Vector Machine, LSSVM)的飞机发动机润滑系统异常状态识别方法。首先对发动机润滑系统参数进行DPCA处理以及在线检测是否有故障发生,如果有故障发生,再采用LSSVM方法进行异常状态识别。以某型飞机发动机润滑系统为例,对文中所提方法的准确性进行试验验证,由试验结果得出文中方法能有效提高飞机发动机异常状态识别准确率。     

A unified classifier for robust face recognition based on combining multiple subspace algorithms

Face recognition being the fastest growing biometric technology has expanded manifold in the last few years. Various new algorithms and commercial systems have been proposed and developed. However, none of the proposed or developed algorithm is a complete solution because it may work very well on one set of images with say illumination changes but may not work properly on another set of image variations like expression variations. This study is motivated by the fact that any single classifier cannot claim to show generally better performance against all facial image variations. To overcome this shortcoming and achieve generality, combining several classifiers using various strategies has been studied extensively also incorporating the question of suitability of any classifier for this task. The study is based on the outcome of a comprehensive comparative analysis conducted on a combination of six subspace extraction algorithms and four distance metrics on three facial databases. The analysis leads to the selection of the most suitable classifiers which performs better on one task or the other. These classifiers are then combined together onto an ensemble classifier by two different strategies of weighted sum and re-ranking. The results of the ensemble classifier show that these strategies can be effectively used to construct a single classifier that can successfully handle varying facial image conditions of illumination, aging and facial expressions.     

Laser-induced breakdown spectroscopy applied to the characterization of rock by support vector machine combined with principal component analysis

Laser-induced breakdown spectroscopy(LIBS) is a versatile tool for both qualitative and quantitative analysis.In this paper,LIBS combined with principal component analysis(PCA) and support vector machine(SVM) is applied to rock analysis.Fourteen emission lines including Fe,Mg,Ca,Al,Si,and Ti are selected as analysis lines.A good accuracy(91.38% for the real rock) is achieved by using SVM to analyze the spectroscopic peak area data which are processed by PCA.It can not only reduce the noise and dimensionality which contributes to improving the efficiency of the program,but also solve the problem of linear inseparability by combining PCA and SVM.By this method,the ability of LIBS to classify rock is validated.     

一种基于PCA和Hough变换的类星体自动识别方法

确定类星体的红移是类星体识别的一个重要目标。由于过去类星体发射谱的静止发射谱模板主要靠天文学家的先验知识得到 ,因此采用模板匹配的方法效果一般不太理想。本文的主要工作有以下二个方面 :(1 )选用了主分量分析方法构造类星体静止发射谱的模板 ,这样 ,模板来源于实际光谱 ,而不是依赖于天文学家的推断 ,因此更加接近实际情况 ;(2 )用二维Hough变换来自动确定红移值。尽管理论上来说 ,只有一个红移参数需要确定 ,但观测发现 ,在大多数情况下发射峰的强度同时发生了变化 ,故在Hough变换中加入强度比例因子作为另一个参数 ,这样可有效提高正确识别率。实验证明 ,本文所提出的方法是可行的 ,正确识别率可达 90 %左右。     

小波重构与局部DCT的二维主元分析掌纹识别

为了解决掌纹特征提取过程中,手掌的非平面问题导致的伪主线噪音信息并简化运算,对重构的掌纹图像进行了局部离散余弦变换.解决了重构后图像噪音敏感性问题,有效地区分了掌纹主线、掌纹褶纹和乳突纹.由二维主元分析算法获得较稳健的识别特征.通过香港理工大学公布的PolyU掌纹数据库的实验,同二维主元分析算法相比,小波重构与局部离散余弦变换的2DPCA掌纹识别算法正确识别率较高,识别效率较高.     

基于FastICA的高光谱图像目标分割

针对高光谱图像目标识别与分类的应用背景,提出了一种基于快速独立成分分析的高光谱图像目标分割算法.通过引入虚拟维数对图像中的目标端元数量进行估计,利用基于非监督正交子空间投影的异常端元提取算法自动获取目标端元光谱,并将其作为快速独立成分分析的初始混合矩阵.采用最小噪声分量变换对原始数据进行降维,利用快速独立成分分析从降维后的主成分中依次提取出图像中的独立分量.最后,对各独立分量进行恒虚警率检测与形态学滤波,从而得到最终的目标分割结果.对AVIRIS型高光谱图像的实验结果表明,该方法可有效探测出图像中的目标,并可获得较好的分割结果.     

人脸图像光照预处理的一种新方法

人脸识别是图像分析和理解领域中最成功的应用之一,近年来得到了迅速的发展,但是阻碍人脸识别技术应用到实际中的瓶颈之一——光照问题,一直没能得到很好的解决。局部二值模式是最近发展起来的一种理论简单但功能强大的纹理分析算法,在计算机视觉等领域表现出良好的性能。将该纹理提取算法应用到图像预处理中并并利用大规模中国人脸图像数据库CAS-PEAL-R1来检验这种方法的有效性。实验结果表明:加入LBP纹理后,该方法能较好解决光照变化问题,提高识别性能。     

Electromagnetic interference shielding characteristics of fabric complexes coated with conductive polypyrrole and thermally evaporated Ag

Through the chemical coating of polypyrrole (PPy) doped with naphthalene sulfonic acid (NSA) on electrically insulating poly (ethylene terephthalate) (PET) woven fabric, PPy–NSA/PET complexes were synthesized. By using the electrochemical coating of PPy doped anthraquinone-2-sulfonic acid (AQSA) on PPy–NSA/PET complexes, PPy–AQSA/PPy–NSA/PET complexes were synthesized. The silver (Ag) was thermally vacuum evaporated on the surface of PPy–AQSA/PPy–NSA/PET complexes (Ag|PPy–AQSA/PPy–NSA/PET). Electromagnetic interference (EMI) shielding efficiency (SE) and dc conductivity (σ_dc) of fabric complexes were measured for EMI shielding characteristics and theoretical simulation. The measurement of EMI SE in the frequency range from 50 MHz to 1.5 GHz was performed by using ASTM D4935-99 method. The EMI shielding characteristics such as transmittance, reflectance and absorbance were obtained from the S (scattering)-parameter analysis. We control the contribution of the absorbance or the reflectance to total EMI SE through the coating of conductive PPy and the evaporation Ag.     

Nanopatterned surface with adjustable area coverage and feature size fabricated by photocatalysis

We report an effective approach to fabricate nanopatterns of alkylsilane self-assembly monolayers (SAMs) with desirable coverage and feature size by gradient photocatalysis in TiO₂ aqueous suspension. Growth and photocatalytic degradation of octadecyltrichlorosilane (OTS) were combined to fabricate adjustable monolayered nanopatterns on mica sheet in this work. Systematic atomic force microscopy (AFM) analysis showed that OTS-SAMs that have similar area coverage with different feature sizes and similar feature size with different area coverages can be fabricated by this approach. Contact angle measurement was applied to confirm the gradually varied nanopatterns contributed to the gradient of UV light illumination. Since this approach is feasible for various organic SAMs and substrates, a versatile method was presented to prepare tunable nanopatterns with desirable area coverage and feature size in many applications, such as molecular and biomolecular recognition, sensor and electrode modification.     

一种基于主分量分析的恒星光谱快速分类法

恒星光谱分类是天体光谱自动识别中的重要组成部分。本文主要介绍一种实用的基于主分量分析(PCA)法对恒星光谱进行快速自动的分类方法。该方法在恒星的主分量空间中对样本点进行投影 ,并利用最近邻分类器进行分类 ,获得与恒星MK分类标准的光谱型基本一致的结果。本文的主要工作有 :(1 )利用PCA方法构造恒星光谱的特征矩阵 ,建构恒星的主分量空间 ;(2 )对恒星光谱进行主分量投影 ,对投影点进行光谱型和光度级的分类器设计 ,利用最近邻法分类 ,最后得出恒星的分类树。该分类法速度快 ,分类准确率较高 ,对目前许多大型光谱巡天计划所获得的大量光谱数据的处理有着重要的意义。     

Principal component analysis of non-uniformity in electrical characteristics of HgCdTe photodiode arrays

We present a method of analyzing the non-uniformity in electrical characteristics of HgCdTe photodiode arrays for infrared imaging applications. We have selected dynamic resistance–voltage (R–V) characteristics for analyzing electrical behavior of HgCdTe photodiodes because the dynamic resistance at a given operating voltage directly governs the imager performance and being derivative of current–voltage (I–V) characteristics, it has little impact of the constant shifts due to stray illumination during dark measurements, relaxing the stringent requirement of perfect dark conditions to some extent for performance analysis. We have demonstrated that by using statistical analysis such as correlation of the selected signatures and their principal component analysis, we can identify the root cause of the high non-uniformity among sensor pixels in the array. The method has been implemented using theoretical I–V model of MWIR HgCdTe photodiodes, but it is generic and may be implemented on any other types of diode arrays for theoretical or experimental analysis of their non-uniformity.     

基于GLS模型的光照鲁棒性人脸的识别

针对线性子空间模型在处理具有阴影的人脸图像时出现的不足之处,提出了GLS模型,并将其用于不同光照下的人脸识别。按照测试图像与正确模型间距离尽可能小的原则构造了一个确定最优分组数和子空间维数的标准;采用SVD方法和K平均聚类法将像素分组,并确定每个分组的线性子空间模型;计算测试图像到每个GLS模型中所有分组的线性子空间模型的距离之和,进而识别人脸图像。经假设检验统计表明,基于该模型的方法在处理不同光照下的人脸图像时,效果明显优于其他方法。     

基于RBF神经网络的较低浓度下同步荧光光谱的溢油鉴别

针对海面溢油样品的含量难以确定,同时考虑到海水掺杂及风化等问题的影响,提出了在较低非线性浓度范围内采集溢油嫌疑样品的同步荧光光谱,获取其训练样本集,利用主成分分析法(Principal com-ponent analysis,PCA)提取其特征光谱,结合径向基函数(Radial basis function,RBF)神经网络对肇事样本和嫌疑样本进行模式识别的方法。通过对相近油源原油样品分类识别研究表明:该方法仅需单次对肇事样本同步光谱测量,再借助数据分析,就可以很好区分相近油源溢油样品,外扰对识别率影响也不大。RBF神经网络算法识别率在92%左右。该结论对海洋环境中溢油的实时检测及油指纹数据信息库的建立有重要意义。