基于小波变换和离散余弦变换的fisher脸识别

本文结合几种现有的人脸识别特征提取算法,先对人脸图像进行小波分解去噪;然后通过离散余弦变换对低频分量作进一步特征提取和压缩,保留人脸图像中对光照、姿态、表情变化不敏感的识别信息;接着利用PCA和LDA相结合得到最终的识别特征;最后采用欧式距离和最近邻分类器识别人脸。实验采用ORL标准人脸库验证了这种组合的有效性。     

KTPⅡ类临界相位匹配腔内和频593.5nm激光器

报道了一种光纤耦合激光二极管阵列(LDA)泵浦Nd YVO4 晶体、Ⅱ类临界相位匹配KTP、腔内和频连续波输出的全固态黄光激光器,输出波长为5 93 .5nm的黄激光是由Nd YVO4 晶体的10 64nm和13 42nm谱线和频产生。采用一个线性平凹腔结构,在8W泵浦功率下,获得了最高功率为2 10mW连续波TEM0 0 的低噪声黄激光输出,光光转换效率为2 .6% ,光束质量因子M2 <1.2。     

LDA抽运Nd:YAG/LBO单通道腔内和频589nm激光器

报道了一种光纤耦合激光二极管阵列(LDA)抽运Nd:YAG晶体、腔内I类临界相位匹配LBO和频、连续波输出的全固态589nm激光器的设计和实验结果。黄光激光是由Nd:YAG 晶体的1064nm和1319nm谱线腔内和频产生,其对应能级跃迁分别为4F3/2到4I11/2和4F3／2 到4I13/2。实验采用三镜腔结构,在808nm 12W的抽运功率下,获得了最高功率为384mW连续波TEM00的589nm黄光激光输出,光束质量因子M2<1．2,4h功率不稳定度小于±2％。实验结果表明采用三镜腔进行腔内和频是获得589nm黄光激光的有效方法,并可以应用到 Nd:YAG晶体的其它谱线或具有多条谱线的其它激光增益介质,获得更多不同波长激光输出。     

基于改进主题分布特征的神经网络语言模型

在递归神经网络(RNN)语言模型输入中增加表示当前词所对应主题的特征向量是一种有效利用长时间跨度历史信息的方法。由于在不同文档中各主题的概率分布通常差别很大,该文提出一种使用文档主题概率改进当前词主题特征的方法,并将改进后的特征应用于基于长短时记忆(LSTM)单元的递归神经网络语言模型中。实验表明,在PTB数据集上该文提出的方法使语言模型的困惑度相对于基线系统下降11.8%。在SWBD数据集多候选重估实验中,该文提出的特征使LSTM模型相对于基线模型词错误率(WER)相对下降6.0%;在WSJ数据集上的实验中,该特征使LSTM模型相对于基线模型词错误率(WER)相对下降6.8%,并且在eval92测试集上,改进隐含狄利克雷分布(LDA)特征使RNN效果与LSTM相当。     

MS/MS-Based Molecular Networking: An Efficient Approach for Natural Products Dereplication

Natural products (NPs) have historically played a primary role in the discovery of small-molecule drugs. However, due to the advent of other methodologies and the drawbacks of NPs, the pharmaceutical industry has largely declined in interest regarding the screening of new drugs from NPs since 2000. There are many technical bottlenecks to quickly obtaining new bioactive NPs on a large scale, which has made NP-based drug discovery very time-consuming, and the first thorny problem faced by researchers is how to dereplicate NPs from crude extracts. Remarkably, with the rapid development of omics, analytical instrumentation, and artificial intelligence technology, in 2012, an efficient approach, known as tandem mass spectrometry (MS/MS)-based molecular networking (MN) analysis, was developed to avoid the rediscovery of known compounds from the complex natural mixtures. Then, in the past decade, based on the classical MN (CLMN), feature-based MN (FBMN), ion identity MN (IIMN), building blocks-based molecular network (BBMN), substructure-based MN (MS2LDA), and bioactivity-based MN (BMN) methods have been presented. In this paper, we review the basic principles, general workflow, and application examples of the methods mentioned above, to further the research and applications of these methods.     

Discrimination of coatings on wooden materials using the gas sensor system

The applicability of sensor system for the discrimination of sources of indoor pollution was investigated. As examples of indoor pollution sources, paint and lacquer coatings were considered. Commercially available preparations: Akrylux, Doamlux, Bejca and White Scandinavian were selected for headspace measurements using TGS sensor array. Following issues were investigated: (1) discrimination between water- and solvent-based coatings, (2) discrimination between one component coatings, and (3) discrimination between one component and two component coatings. Following data analysis methods were used: principal component analysis (PCA), linear discriminant analysis (LDA) and probabilistic neural network (PNN). Results showed that coatings could be discriminated successfully, provided the surface covered was solid wood (0-1.8% error). The interference of fibreboard volatiles in sensor measurements of coatings was most likely encountered. It could have significantly impaired discrimination of coatings on fibreboard (2.8-5.6% error) as compared to wood. Worst results were obtained for the discrimination of coatings on unknown material(12.5-28.7% error).     

基于LDA主题模型的情感分析研究

LDA主题模型在提取特征时缺乏对词语关联及相关词对的理解,这会影响情感极性分类的准确率。针对这一问题,文中提出一种在LDA主题模型中引入特征情感词对抽取方法的新模型,以改善特征情感词对的抽取效果。利用依存句法分析设计特征情感词对的识别方法,随后将识别方法作为约束条件引入LDA模型对特征情感词对进行抽取。通过吉布斯采样进行参数计算,给出了模型的生成过程。最后利用随机森林分类方法对文本进行情感极性分类。为验证文中模型的有效性,将其和另外两种模型一起进行实验,当主题个数为20时,文中所提模型分类的准确率、召回率、F值分别为81.54%、83.13%和82.33%,显著高于另外两种模型。     

3T‐FASDM: Linear discriminant analysis‐based three‐tier face anti‐spoofing detection model using support vector machine

In recent years, to solve the problem of face spoofing, momentous work has been done in this field, but still, there is a need for establishing counter measures to the biometric spoofing attacks. Although trained and evaluated on different databases, impressive results have been achieved in existing face anti‐spoofing techniques, but biometric authentication is a very significant problem as imposters are using lots of reconstructed samples or fake synthetic material or structure that can be used for various attack purposes. For the first time, to the best of our knowledge, this paper explains the security for face anti‐spoofing detection using linear discriminant analysis and validates the results by calculating HTER and accuracy on different databases (i.e., REPLAY ATTACK and CASIA). The proposed model, that is, three‐tier face anti‐spoofing detection model (3T‐FASDM), is used for the detection of the fake biometric user and works well for real‐time applications. The proposed methods tested on a set of state‐of‐the‐art anti‐spoofing features for the face mode gives a very low degree of complexity as 26 general image quality measures are applied to differentiate among legitimate and imposter samples. The outcomes obtained from publically available data show that this technique has improved performance and accuracy by analyzing the HTER and machine learning classifiers that are helpful to differentiate among real and fake traits.     

Usefulness of chemometrics and mass spectrometry-based electronic nose to classify Australian white wines by their varietal origin

A combination of mass spectrometry-based electronic nose (MS e_nose) and chemometrics was explored to classify two Australian white wines according to their varietal origin namely Riesling and unwooded Chardonnay. The MS e_nose data were analysed using principal components analysis (PCA), discriminant partial least squares (DPLS) and linear discriminant analysis (LDA) applied to principal components scores and validated using full cross validation (leave one out). DPLS gave the highest levels of correct classification for both varieties (>90%). LDA classified correctly 73% of unwooded Chardonnay and 82% of Riesling wines. Even though the conventional analysis provides fundamental information about the volatile compounds present in the wine, the MS e_nose method has a series of advantages over conventional analytical techniques due to simplicity of the sample-preparation and reduced time of analysis and might be considered as a more convenient choice for routine process control in an industrial environment. The work reported here is a feasibility study and requires further development with considerably more commercial samples of different varieties. Further studies are needed in order to improve the calibration specificity, accuracy and robustness, and to extend the discrimination to other wine varieties or blends.     

Effect of the aryl group substituent in the dimerization of 3-arylisoxazoles to syn 2,6-diaryl-3,7-diazatricyclo[4.2.0.0]octan-4,8-diones induced by LDA

3-Arylisoxazoles react with LDA in THF at 0 °C affording syn-2,6-diaryl-3,7-diazatricyclo[4.2.0.0^2,5]octan-4,8-diones (bis-azetidinones), via stereoselective dimerization of an azetinone anion intermediate. A fragmentation reaction affording arylnitriles may compete with electronic and steric effects of the substituent present in the aryl group being pivotal in determining the outcome of this reaction. An interesting behaviour with LDA of arylnitriles arising from the fragmentation reaction of some 3-arylisoxazoles was also observed. N,N-Diisopropylaminobenzonitriles were in fact formed (plausibly via a benzyne mechanism) from 3-(4-chlorophenyl)isoxazole and 3-(2-chlorophenyl)isoxazole, whereas 3-(2-methylphenyl)isoquinolin-1-amine was isolated starting from 3-(2-methylphenyl)isoxazole and LDA.