The influence of Impedance Boundary Condition (IBC) on transonic compressors is investigated. A systematic input–output analytical framework is developed, which treats the nonlinearities as unknown forcing terms. The...The influence of Impedance Boundary Condition (IBC) on transonic compressors is investigated. A systematic input–output analytical framework is developed, which treats the nonlinearities as unknown forcing terms. The framework is validated through the experiments of rotating inlet distortion within a low-speed compressor. The input–output method is subsequently applied to transonic compressors, including NASA Rotor37 and Stage35, wherein impedance optimization is studied along with the exploration of its fundamental mechanisms. The IBC is employed to model the effect of Casing Treatment (CT). The optimal complex impedance values are determined through predicted results and tested across a range of circumferential modes and forcing frequencies. The IBC significantly reduces the energy and Reynolds stress gain, notably at the first-order circumferential mode and within the Rotor Rotating Frequency (RRF) range. Output modes reveal that transonic compressors with fine-tuned impedance values exhibit a more confined perturbation distribution and redistribute the perturbations compared to the uncontrolled case. Additionally, the roles of resistance and reactance are elucidated through input–output analysis, and resistance determines the energy transfer direction between flow and pressure waves and modulates the amplitude, whereas reactance modifies the phase relationships and attenuates the perturbations.展开更多
In principal component analysis （PCA） algorithms for face recognition, to reduce the influence of the eigenvectors which relate to the changes of the illumination on abstract features, a modified PCA （MPCA） algori...In principal component analysis （PCA） algorithms for face recognition, to reduce the influence of the eigenvectors which relate to the changes of the illumination on abstract features, a modified PCA （MPCA） algorithm is proposed. The method is based on the idea of reducing the influence of the eigenvectors associated with the large eigenvalues by normalizing the feature vector element by its corresponding standard deviation. The Yale face database and Yale face database B are used to verify the method. The simulation results show that, for front face and even under the condition of limited variation in the facial poses, the proposed method results in better performance than the conventional PCA and linear discriminant analysis （LDA） approaches, and the computational cost remains the same as that of the PCA, and much less than that of the LDA.展开更多
Highly toxic phosgene,diethyl chlorophosphate(DCP)and volatile acyl chlorides endanger our life and public security.To achieve facile sensing and discrimination of multiple target analytes,herein,we presented a single...Highly toxic phosgene,diethyl chlorophosphate(DCP)and volatile acyl chlorides endanger our life and public security.To achieve facile sensing and discrimination of multiple target analytes,herein,we presented a single fluorescent probe(BDP-CHD)for high-throughput screening of phosgene,DCP and volatile acyl chlorides.The probe underwent a covalent cascade reaction with phosgene to form boron dipyrromethene(BODIPY)with bright green fluorescence.By contrast,DCP,diphosgene and acyl chlorides can covalently assembled with the probe,giving rise to strong blue fluorescence.The probe has demonstrated high-throughput detection capability,high sensitivity,fast response(within 3 s)and parts per trillion(ppt)level detection limit.Furthermore,a portable platform based on BDP-CHD was constructed,which has achieved high-throughput discrimination of 16 analytes through linear discriminant analysis(LDA).Moreover,a smartphone adaptable RGB recognition pattern was established for the quantitative detection of multi-analytes.Therefore,this portable fluorescence sensing platform can serve as a versatile tool for rapid and high-throughput detection of toxic phosgene,DCP and volatile acyl chlorides.The proposed“one for more”strategy simplifies multi-target discrimination procedures and holds great promise for various sensing applications.展开更多
Strong sensitivity of satellite microwave remote sensing to the change of surface dielectric properties,as well as the insensitivity to air pollution and solar illumination effects,makes it very suitable for monitorin...Strong sensitivity of satellite microwave remote sensing to the change of surface dielectric properties,as well as the insensitivity to air pollution and solar illumination effects,makes it very suitable for monitoring freeze-thaw conditions.The freeze-thaw cycle changes in the Qinghai-Xizang Plateau have an important impact on the ecological environment and infrastructure.Based on the Scanning Multi-channel Microwave Radiometer(SMMR)and other sensors of microwave satellite,the freeze-thaw cycle data of permafrost in the Qinghai-Xizang Plateau in the past 40 years from 1981 to 2020 was obtained.The changes of soil freeze-thaw conditions in different seasons of 2020 and in the same season of 1990,2000,2010 and 2020 were compared,and the annual variation trend of soil freeze-thaw area in the four years was analyzed.Further,the linear regression analysis was carried out on the duration of soil freezing/thawing/transition and the interannual variation trend under different area conditions from 1981 to 2020.The results show that the freeze-thaw changes in different years are similar.In winter,it is mainly frozen for about 110 days.Spring and autumn are transitional periods,lasting for 170 days.In summer,it is mainly thawed for about 80 days.From 1981 to 2020,the freezing period and the average freezing area of the Qinghai-Xizang Plateau decreased at a rate of 0.22 days and 1986 km^(2) per year,respectively,while the thawing period and the average thawing area increased at a rate of 0.07 days and 3187 km^(2) per year,respectively.The research results provide important theoretical support for the ecological environment and permafrost protection of the Qinghai-Xizang Plateau.展开更多
Nitrogen(N)is the most important nutrient for plants;however,microbe-mediated N transformation under different N forms is unclear.This experiment investigated the effects of four treatments fertilized with various N f...Nitrogen(N)is the most important nutrient for plants;however,microbe-mediated N transformation under different N forms is unclear.This experiment investigated the effects of four treatments fertilized with various N forms,no N(control,CK),100%ammonium N(AN),100%nitrate N(NN),and 50%ammonium N+50%nitrate N(ANNN),on soil chemical properties,rhizosphere bacterial network,and rice growth.The ANNN treatment enhanced soil pH by 6.9%,soil organic carbon by 12%,and microbial biomass N(MBN)by 60%compared to CK.The linear discriminant effect size(LEfSe)analysis indicated four highly abundant biomarkers of bacterial communities each in the CK,NN,and AN treatments,while the ANNN treatment showed six highly abundant biomarkers with maximum effect size and linear discriminant analysis(LDA)score>4.The 16S rRNA gene-predicted functions under PICRUST indicated glutathione metabolism and proteasome and Tax4Fun recorded amino acid metabolism in the ANNN treatment.The combination of ammonium and nitrate N(i.e.,the ANNN treatment)significantly increased the expression levels of the genes encoding N metabolism,including AMT1,NRT2.1,GS1,and GOGAT1,and induced 39%,27%,35%,and 38%increase in nitrate reductase,nitrite reductase,glutamine synthetase,and glutamate synthase,respectively,in comparison to CK.In addition,the ANNN treatment promoted rice leaf photosynthetic rate by 37%,transpiration rate by 41%,CO_(2) exchange rate by 11%,and stomatal conductance by 18%compared to CK,while increased N use efficiency(NUE)by 10%and 19%,respectively,compared to the AN and NN treatments.These findings suggest that the combination of ammonium and nitrate N can promote bacterial community abundance,composition,and functional pathways by improving soil properties and can increase NUE and rice growth.This study provides a theoretical basis for the rational application of N fertilizers and the implications of this approach for future sustainable crop production.展开更多
The dysbiosis of oral microbiota contributes to diseases such as periodontitis and certain cancers by triggering the host inflammatory response.Developing methods for the immediate and sensitive identification of oral...The dysbiosis of oral microbiota contributes to diseases such as periodontitis and certain cancers by triggering the host inflammatory response.Developing methods for the immediate and sensitive identification of oral microorganism is crucial for the rapid diagnosis and early interventions of associated diseases.Traditional methods for microbial detection primarily include the plate culturing,polymerase chain reaction and enzyme-linked immunosorbent assay,which are either time-consuming or laborious.Herein,we reported a persistent luminescence-encoded multiple-channel optical sensing array and achieved the rapid and accurate identification of oral-derived microorganisms.Our results demonstrate that electrostatic attractions and hydrophobic-hydrophobic interactions dominate the binding of the persistent luminescent nanoprobes to oral microorganisms and the microbial identification process can be finished within 30 min.Specifically,a total of 7 oral-derived microorganisms demonstrate their own response patterns and were differentiated by linear discriminant analysis(LDA)with the accuracy up to 100%both in the solution and artificial saliva samples.Moreover,the persistent luminescence encoded array sensor could also discern the microorganism mixtures with the accuracy up to 100%.The proposed persistent luminescence encoding sensor arrays in this work might offer new ideas for rapid and accurate oralderived microorganism detection,and provide new ways for disease diagnosis associated with microbial metabolism.展开更多
We study the existence and stability of dark-gap solitons in linear lattice and nonlinear lattices.The results indicate that the combination of linear and nonlinear lattices gives dark-gap solitons unique properties.T...We study the existence and stability of dark-gap solitons in linear lattice and nonlinear lattices.The results indicate that the combination of linear and nonlinear lattices gives dark-gap solitons unique properties.The linear lattice can stabilize dark-gap solitons,while the nonlinear lattice reduces the stability of dark-gap solitons.On the basis of numerical analysis,we investigate the effects of lattice depth,chemical potential,nonlinear lattice amplitude,and nonlinear lattice period on the soliton in mixed lattices with the same and different periods.The stability of dark-gap soliton is studied carefully by means of real-time evolution and linear stability analysis.Dark-gap solitons can exist stably in the band gap,but the solitons formed by the mixed lattices are slightly different when the period is the same or different.展开更多
Alternate path(AP) method is the most widely used method for the progressive collapse analysis,and its application in frame structures has been well proved.However,the application of AP method for other structures,esp...Alternate path(AP) method is the most widely used method for the progressive collapse analysis,and its application in frame structures has been well proved.However,the application of AP method for other structures,especially for cable-stayed structures,should be further developed.The four analytical procedures,i.e.,linear static,nonlinear static,linear dynamic,and nonlinear dynamic were firstly improved by taking into account the initial state.Then a cable-stayed structure was studied using the four improved methods.Furthermore,the losses of both one cable and two cables were discussed.The results show that for static and dynamic analyses of the cable-stayed bridges,there is large difference between the results obtained from simulations starting with either a deformed or a nondeformed configuration at the time of cable loss.The static results are conservative in the vicinity of the ruptured cable,but the dynamic effect of the cable loss in the area farther away from the loss-cable cannot be considered.Moreover,the dynamic amplification factor of 2.0 is found to be a good estimate for static analysis procedures,since linear static and linear dynamic procedures yield approximately the same maximum vertical deflection.The results of the comprehensive evaluation of the cable failure show that the tread of the progressive failure of the cable-stayed bridges decreases when the location of the failed cables is closer to the pylon.展开更多
A direct linear discriminant analysis algorithm based on economic singular value decomposition （DLDA/ESVD） is proposed to address the computationally complex problem of the conventional DLDA algorithm, which directl...A direct linear discriminant analysis algorithm based on economic singular value decomposition （DLDA/ESVD） is proposed to address the computationally complex problem of the conventional DLDA algorithm, which directly uses ESVD to reduce dimension and extract eigenvectors corresponding to nonzero eigenvalues. Then a DLDA algorithm based on column pivoting orthogonal triangular （QR） decomposition and ESVD （DLDA/QR-ESVD） is proposed to improve the performance of the DLDA/ESVD algorithm by processing a high-dimensional low rank matrix, which uses column pivoting QR decomposition to reduce dimension and ESVD to extract eigenvectors corresponding to nonzero eigenvalues. The experimental results on ORL, FERET and YALE face databases show that the proposed two algorithms can achieve almost the same performance and outperform the conventional DLDA algorithm in terms of computational complexity and training time. In addition, the experimental results on random data matrices show that the DLDA/QR-ESVD algorithm achieves better performance than the DLDA/ESVD algorithm by processing high-dimensional low rank matrices.展开更多
The Stereo Particle Image Velocimetry(SPIV)technology is applied to measure the wingtip vortices generated by the up-down symmetrical split winglet.Then,the temporal biglobal Linear Stability Analysis(bi-global LSA)is...The Stereo Particle Image Velocimetry(SPIV)technology is applied to measure the wingtip vortices generated by the up-down symmetrical split winglet.Then,the temporal biglobal Linear Stability Analysis(bi-global LSA)is performed on this nearly equal-strength corotating vortex pair,which is composed of an upper vortex(vortex-u)and a down vortex(vortex-d).The results show that the instability eigenvalue spectrum illustrated by(ωr,ω_(i))contains two types of branches:discrete branch and continuous branch.The discrete branch contains the primary branches of vortex-u and vortex-d,the secondary branch of vortex-d and coupled branch,of which all of the eigenvalues are located in the unstable half-plane ofω_(i)>0,indicating that the wingtip vortex pair is temporally unstable.By contrast,the eigenvalues of the continuous branch are concentrated on the half-plane ofω_(i)<0 and the perturbation modes correspond to the freestream perturbation.In the primary branches of vortex-u and vortex-d,Mode P_(u) and Mode Pd are the primary perturbation modes,which exhibit the structures enclosed with azimuthal wavenumber m and radial wavenumber n,respectively.Besides,the results of stability curves for vortex-u and vortex-d demonstrate that the instability growth rates of vortex-u are larger than those of vortex-d,and the perturbation energy of Mode P_(u) is also larger than that of Mode Pd.Moreover,the perturbation energy of Mode P_(u) is up to 0.02650 and accounts for 33.56%percent in the corresponding branch,thereby indicating that the instability development of wingtip vortex is dominated by Mode P_(u).By further investigating the topological structures of Mode P_(u) and Mode Pd with streamwise wavenumbers,the most unstable perturbation mode with a large azimuthal wavenumber of m=5-6 is identified,which imposes on the entire core region of vortex-u.This large azimuthal wavenumber perturbation mode can suggest the potential physical-based flow control strategy by manipulating it.展开更多
Four samples from a Permian reservoir in the Ordos Basin of North China were separated into twelve fractions in grain sizes of 〈0.5, 0.5-1 and 1-2 μm. Using the ^40Ar-^39Ar step-heating method, all of the fractions ...Four samples from a Permian reservoir in the Ordos Basin of North China were separated into twelve fractions in grain sizes of 〈0.5, 0.5-1 and 1-2 μm. Using the ^40Ar-^39Ar step-heating method, all of the fractions essentially yielded plateau ages ranging from 172.5 to 217.1 Ma. These scattered plateau ages might not have been obtained from pure diagenetic illites but from mixed clay minerals, although the samples were disaggregated using a gentle freeze-thaw cycle to free them of non-clay minerals. A regional thermal event, as suggested by several proxies, led to intensive iHitization as a distinct diagenetic process when the Yanshanian Movement triggered magmatism around the entire North China Block during the Jurassic to Cretaceous. Thermal illites formed during a short time period, whereas detrital illites came from various sources. The scattered plateau ages could have resulted from mixed degassing of thermal and detrital illites. Within one sample, the plateau ages decrease with the diminution of grain sizes, but it is difficult to extrapolate to the detrital-illite-free ages. Because the plateau age is a mixture of ages for thermal and detrital illites, this regression analysis studies the dependence of the plateau ages on the synthetic values of contents and ages of detrital illites instead of on the grain sizes. Comparing the samples to one another, the plateau ages show the same trend among the different grain sizes. Weighted by the contents and ages of detrital illites, linear regression analysis revealed the relationship between the plateau ages and the relative weight proportions. Based on iterated calculations, a thermal event age and a set of weight proportions were derived. The regressed thermal event age is 163.3±1.6 Ma, which coincides with regional thermal activities and links to gas accumulation.展开更多
Vehicle traveling time prediction is an important part of the research of intelligent transportation system. By now, there have been various kinds of methods for vehicle traveling time prediction. But few consider bot...Vehicle traveling time prediction is an important part of the research of intelligent transportation system. By now, there have been various kinds of methods for vehicle traveling time prediction. But few consider both aspects of time and space. In this paper, a vehicle traveling time prediction method based on grey theory （GT） and linear regression analysis （LRA） is presented. In aspects of time, we use the history data sequence of bus speed on a certain road to predict the future bus speed on that road by GT. And in aspects of space, we calculate the traffic affecting factors between various roads by LRA. Using these factors we can predict the vehicle＇s speed at the lower road if the vehicle＇s speed at the current road is known. Finally we use time factor and space factor as the weighting factors of the two results predicted by GT and LRA respectively to find the final result, thus calculating the vehicle＇s traveling time. The method also considers such factors as dwell time, thus making the prediction more accurate.展开更多
A series of linear stability analysis is carried out on the onset of thermal convection in the presence of spatial variations of viscosity, thermal conductivity and expansivity. We consider the temporal evolution of a...A series of linear stability analysis is carried out on the onset of thermal convection in the presence of spatial variations of viscosity, thermal conductivity and expansivity. We consider the temporal evolution of an infinitesimal perturbation superimposed to a static （motionless） and con- ductive state in a basally-heated planar layer. From the changes in flow patterns with increasing the amplitudes of temperature dependence of viscosity, we identified the transition into the ＂stagnant-lid＂ （ST） regime, where the convection occurs only beneath a thick and stagnant-lid of cold fluid at the top surface. Detailed analysis showed a significant increase of the aspect ratio of convection cells in ST regime induced by the spatial variations in thermal conductivity and/or expansivity： the horizon- tal length scale of ST convection can be enlarged by up to 50% with 10 times increase of thermal conductivity with depth. We further developed an analytical model of ST convection which success- fully reproduced the mechanism of increasing horizontal length scale of ST regime convection cells for given spatial variations in physical properties. Our findings may highlight the essential roles of the spatial variation of thermal conductivity on the convection patterns in the mantle.展开更多
Security is a vital parameter to conserve energy in wireless sensor networks(WSN).Trust management in the WSN is a crucial process as trust is utilized when collaboration is important for accomplishing trustworthy dat...Security is a vital parameter to conserve energy in wireless sensor networks(WSN).Trust management in the WSN is a crucial process as trust is utilized when collaboration is important for accomplishing trustworthy data transmission.But the available routing techniques do not involve security in the design of routing techniques.This study develops a novel statistical analysis with dingo optimizer enabled reliable routing scheme(SADO-RRS)for WSN.The proposed SADO-RRS technique aims to detect the existence of attacks and optimal routes in WSN.In addition,the presented SADORRS technique derives a new statistics based linear discriminant analysis(LDA)for attack detection,Moreover,a trust based dingo optimizer(TBDO)algorithm is applied for optimal route selection in the WSN and accomplishes secure data transmission in WSN.Besides,the TBDO algorithm involves the derivation of the fitness function involving different input variables of WSN.For demonstrating the enhanced outcomes of the SADO-RRS technique,a wide range of simulations was carried out and the outcomes demonstrated the enhanced outcomes of the SADO-RRS technique.展开更多
Visual process monitoring is important in complex chemical processes.To address the high state separation of industrial data,we propose a new criterion for feature extraction called balanced multiple weighted linear d...Visual process monitoring is important in complex chemical processes.To address the high state separation of industrial data,we propose a new criterion for feature extraction called balanced multiple weighted linear discriminant analysis(BMWLDA).Then,we combine BMWLDA with self-organizing map(SOM)for visual monitoring of industrial operation processes.BMWLDA can extract the discriminative feature vectors from the original industrial data and maximally separate industrial operation states in the space spanned by these discriminative feature vectors.When the discriminative feature vectors are used as the input to SOM,the training result of SOM can differentiate industrial operation states clearly.This function improves the performance of visual monitoring.Continuous stirred tank reactor is used to verify that the class separation performance of BMWLDA is more effective than that of traditional linear discriminant analysis,approximate pairwise accuracy criterion,max–min distance analysis,maximum margin criterion,and local Fisher discriminant analysis.In addition,the method that combines BMWLDA with SOM can effectively perform visual process monitoring in real time.展开更多
The moving window bis corelation coefficients(MW BiCC)was proposed and employed for the discriminant analysis of transgenic sugarcane leaves and B-thalassemia with visible and near-infrared(Vis NIR)spectroscopy.The we...The moving window bis corelation coefficients(MW BiCC)was proposed and employed for the discriminant analysis of transgenic sugarcane leaves and B-thalassemia with visible and near-infrared(Vis NIR)spectroscopy.The well-performed moving window principal component analysis linear discriminant analysis(MWPCA-LDA)was also conducted for comparison.A total of 306 transgenic(positive)and 150 nont ransgenic(negative)leave samples of sugarcane were collected and divided to calibration,prediction,and validation.The diffuse reflection spectra were corected using Savitzky-Golay(SG)smoothing with first-order derivative(d=1),third-degree polynomial(p=3)and 25 smpothing points(m=25).The selected waveband was 736-1054nm with MW-BiCC,and the positive and negative validation recognition rates(V_REC^(+),VREC^(-))were 100%,98.0%,which achieved the same effect as MWPCA-LDA.Another example,the 93 B-thalassemia(positive)and 148 nonthalassemia(negative)of human hemolytic samples were colloctod.The transmission spectra were corrected using SG smoothing withd=1,p=3 and m=53.Using M W-BiCC,many best wavebands were selected(e.g.,1116-1146,17941848 and 22842342nm).The V_REC^(+)and V_REC^(-)were both 100%,which achieved the same effect as MW-PCA-LDA.Importantly,the BICC only required ca lculating correlation cofficients between the spectrum of prediction sample and the average spectra of two types of calibration samples.Thus,BiCC was very simple in algorithm,and expected to obtain more applications.The results first confirmed the feasibility of distinguishing B-thalassemia and normal control samples by NIR spectroscopy,and provided a promising simple tool for large population thalassemia screening.展开更多
This study proposes an approach based on machine learning to forecast currency exchange rates by applying sentiment analysis to messages on Twitter(called tweets).A dataset of the exchange rates between the United Sta...This study proposes an approach based on machine learning to forecast currency exchange rates by applying sentiment analysis to messages on Twitter(called tweets).A dataset of the exchange rates between the United States Dollar(USD)and the Pakistani Rupee(PKR)was formed by collecting information from a forex website as well as a collection of tweets from the business community in Pakistan containing finance-related words.The dataset was collected in raw form,and was subjected to natural language processing by way of data preprocessing.Response variable labeling was then applied to the standardized dataset,where the response variables were divided into two classes:“1”indicated an increase in the exchange rate and“−1”indicated a decrease in it.To better represent the dataset,we used linear discriminant analysis and principal component analysis to visualize the data in three-dimensional vector space.Clusters that were obtained using a sampling approach were then used for data optimization.Five machine learning classifiers—the simple logistic classifier,the random forest,bagging,naïve Bayes,and the support vector machine—were applied to the optimized dataset.The results show that the simple logistic classifier yielded the highest accuracy of 82.14%for the USD and the PKR exchange rates forecasting.展开更多
In terms of formation mechanisms of linear dunes,there are open arguments for their widespread distribution and multi-morphological diversities.In order to clarify the formation mechanism of linear dunes of Qarhan Sal...In terms of formation mechanisms of linear dunes,there are open arguments for their widespread distribution and multi-morphological diversities.In order to clarify the formation mechanism of linear dunes of Qarhan Salt Lake,we used pattern analysis method to analyze the statistical characteristics and spatial variation of their pattern parameters.Except at the west-northwest margin,the pattern parameters showed regular spatial variation from the up-middle part towards the downwind end of the dune field.Based on the cumulative probability plots for inter-crest spacing and crest length,we divided the linear dunes into three groups,which corresponding to the three evolution stages of these dunes.The first group comprises erosional relics,with shorter crests,smaller inter-crest spacing and more random dune orientation.The second group comprises dunes whose sand supply is just sufficient to maintain stability and these dunes are approaching the net erosion stage.The crest length and inter-crest spacing of these dunes are much larger than those of the first group,and dune orientation is closer to the resultant drift direction （RDD） .The last group comprises linear dunes that are still undergoing vertical accretion and longitudinal elongation,which follows the RDD of the modern wind regime.The presence of regular spatial variation of pattern parameters and a similar geometry with the vegetated linear dunes suggest that deposition and erosion coexist in the development and evolution of linear dunes of Qarhan Salt Lake,i.e.deposition predominates at the downwind end of linear dunes in the vertical accretion and longitudinal elongation stage,whereas erosion mainly occurs at the upwind end of linear dunes in the degradation stage.Therefore,the formation mechanism of linear dunes in Qarhan Salt Lake can be reasonably explained by the combination of depositional and erosional theories.展开更多
A kernel-based discriminant analysis method called kernel direct discriminant analysis is employed, which combines the merit of direct linear discriminant analysis with that of kernel trick. In order to demonstrate it...A kernel-based discriminant analysis method called kernel direct discriminant analysis is employed, which combines the merit of direct linear discriminant analysis with that of kernel trick. In order to demonstrate its better robustness to the complex and nonlinear variations of real face images, such as illumination, facial expression, scale and pose variations, experiments are carried out on the Olivetti Research Laboratory, Yale and self-built face databases. The results indicate that in contrast to kernel principal component analysis and kernel linear discriminant analysis, the method can achieve lower （7%） error rate using only a very small set of features. Furthermore, a new corrected kernel model is proposed to improve the recognition performance. Experimental results confirm its superiority （1% in terms of recognition rate） to other polynomial kernel models.展开更多
Linear discrimiant analysis (LDA) has been used in face recognition. But it is difficult to handle the high nonlinear problems, such as changes of large viewpoint and illumination. In order to overcome these problems,...Linear discrimiant analysis (LDA) has been used in face recognition. But it is difficult to handle the high nonlinear problems, such as changes of large viewpoint and illumination. In order to overcome these problems, kernel discriminant analysis for face recognition is presented. This approach adopts the kernel functions to replace the dot products of nonlinear mapping in the high dimensional feature space, and then the nonlinear problem can be solved in the input space conveniently without explicit mapping. Two face databases are given.展开更多
基金co-supported by the National Natural Science Foundation of China(Nos.52325602,52306036 and 52306035)the National Science and Technology Major Project of China(No.Y2022-II-0003-0006 and Y2022-II-0002-0005)+1 种基金the project funded by China Postdoctoral Science Foundation(No.2022M720346)supported by the Key Laboratory of Pre-Research Management Centre of China(No.6142702200101).
文摘The influence of Impedance Boundary Condition (IBC) on transonic compressors is investigated. A systematic input–output analytical framework is developed, which treats the nonlinearities as unknown forcing terms. The framework is validated through the experiments of rotating inlet distortion within a low-speed compressor. The input–output method is subsequently applied to transonic compressors, including NASA Rotor37 and Stage35, wherein impedance optimization is studied along with the exploration of its fundamental mechanisms. The IBC is employed to model the effect of Casing Treatment (CT). The optimal complex impedance values are determined through predicted results and tested across a range of circumferential modes and forcing frequencies. The IBC significantly reduces the energy and Reynolds stress gain, notably at the first-order circumferential mode and within the Rotor Rotating Frequency (RRF) range. Output modes reveal that transonic compressors with fine-tuned impedance values exhibit a more confined perturbation distribution and redistribute the perturbations compared to the uncontrolled case. Additionally, the roles of resistance and reactance are elucidated through input–output analysis, and resistance determines the energy transfer direction between flow and pressure waves and modulates the amplitude, whereas reactance modifies the phase relationships and attenuates the perturbations.
文摘In principal component analysis （PCA） algorithms for face recognition, to reduce the influence of the eigenvectors which relate to the changes of the illumination on abstract features, a modified PCA （MPCA） algorithm is proposed. The method is based on the idea of reducing the influence of the eigenvectors associated with the large eigenvalues by normalizing the feature vector element by its corresponding standard deviation. The Yale face database and Yale face database B are used to verify the method. The simulation results show that, for front face and even under the condition of limited variation in the facial poses, the proposed method results in better performance than the conventional PCA and linear discriminant analysis （LDA） approaches, and the computational cost remains the same as that of the PCA, and much less than that of the LDA.
基金the financial support of the National Natural Science Foundation of China(No.22168009)。
文摘Highly toxic phosgene,diethyl chlorophosphate(DCP)and volatile acyl chlorides endanger our life and public security.To achieve facile sensing and discrimination of multiple target analytes,herein,we presented a single fluorescent probe(BDP-CHD)for high-throughput screening of phosgene,DCP and volatile acyl chlorides.The probe underwent a covalent cascade reaction with phosgene to form boron dipyrromethene(BODIPY)with bright green fluorescence.By contrast,DCP,diphosgene and acyl chlorides can covalently assembled with the probe,giving rise to strong blue fluorescence.The probe has demonstrated high-throughput detection capability,high sensitivity,fast response(within 3 s)and parts per trillion(ppt)level detection limit.Furthermore,a portable platform based on BDP-CHD was constructed,which has achieved high-throughput discrimination of 16 analytes through linear discriminant analysis(LDA).Moreover,a smartphone adaptable RGB recognition pattern was established for the quantitative detection of multi-analytes.Therefore,this portable fluorescence sensing platform can serve as a versatile tool for rapid and high-throughput detection of toxic phosgene,DCP and volatile acyl chlorides.The proposed“one for more”strategy simplifies multi-target discrimination procedures and holds great promise for various sensing applications.
基金National Natural Science foundation of China(No.42271432)Foundation of Shanxi Vocational University of Engineering Science and Technology(No.KJ 202426).
文摘Strong sensitivity of satellite microwave remote sensing to the change of surface dielectric properties,as well as the insensitivity to air pollution and solar illumination effects,makes it very suitable for monitoring freeze-thaw conditions.The freeze-thaw cycle changes in the Qinghai-Xizang Plateau have an important impact on the ecological environment and infrastructure.Based on the Scanning Multi-channel Microwave Radiometer(SMMR)and other sensors of microwave satellite,the freeze-thaw cycle data of permafrost in the Qinghai-Xizang Plateau in the past 40 years from 1981 to 2020 was obtained.The changes of soil freeze-thaw conditions in different seasons of 2020 and in the same season of 1990,2000,2010 and 2020 were compared,and the annual variation trend of soil freeze-thaw area in the four years was analyzed.Further,the linear regression analysis was carried out on the duration of soil freezing/thawing/transition and the interannual variation trend under different area conditions from 1981 to 2020.The results show that the freeze-thaw changes in different years are similar.In winter,it is mainly frozen for about 110 days.Spring and autumn are transitional periods,lasting for 170 days.In summer,it is mainly thawed for about 80 days.From 1981 to 2020,the freezing period and the average freezing area of the Qinghai-Xizang Plateau decreased at a rate of 0.22 days and 1986 km^(2) per year,respectively,while the thawing period and the average thawing area increased at a rate of 0.07 days and 3187 km^(2) per year,respectively.The research results provide important theoretical support for the ecological environment and permafrost protection of the Qinghai-Xizang Plateau.
基金financially supported by the National Natural Science Foundation of China(No.32172109)the Natural Science Foundation of Guangdong Province,China(No.2021A1515010566).
文摘Nitrogen(N)is the most important nutrient for plants;however,microbe-mediated N transformation under different N forms is unclear.This experiment investigated the effects of four treatments fertilized with various N forms,no N(control,CK),100%ammonium N(AN),100%nitrate N(NN),and 50%ammonium N+50%nitrate N(ANNN),on soil chemical properties,rhizosphere bacterial network,and rice growth.The ANNN treatment enhanced soil pH by 6.9%,soil organic carbon by 12%,and microbial biomass N(MBN)by 60%compared to CK.The linear discriminant effect size(LEfSe)analysis indicated four highly abundant biomarkers of bacterial communities each in the CK,NN,and AN treatments,while the ANNN treatment showed six highly abundant biomarkers with maximum effect size and linear discriminant analysis(LDA)score>4.The 16S rRNA gene-predicted functions under PICRUST indicated glutathione metabolism and proteasome and Tax4Fun recorded amino acid metabolism in the ANNN treatment.The combination of ammonium and nitrate N(i.e.,the ANNN treatment)significantly increased the expression levels of the genes encoding N metabolism,including AMT1,NRT2.1,GS1,and GOGAT1,and induced 39%,27%,35%,and 38%increase in nitrate reductase,nitrite reductase,glutamine synthetase,and glutamate synthase,respectively,in comparison to CK.In addition,the ANNN treatment promoted rice leaf photosynthetic rate by 37%,transpiration rate by 41%,CO_(2) exchange rate by 11%,and stomatal conductance by 18%compared to CK,while increased N use efficiency(NUE)by 10%and 19%,respectively,compared to the AN and NN treatments.These findings suggest that the combination of ammonium and nitrate N can promote bacterial community abundance,composition,and functional pathways by improving soil properties and can increase NUE and rice growth.This study provides a theoretical basis for the rational application of N fertilizers and the implications of this approach for future sustainable crop production.
基金financially supported by Quanzhou high-level Talents Project Fund(No.2022C033R)the National Natural Science Foundation of China(Nos.21925401,52221001)+2 种基金the Fundamental Research Funds for the Central Universities(No.2042022rc0004)the Postdoctoral Innovative Research of Hubei Province of China(No.211000025)the interdisciplinary innovative talents foundation from Renmin Hospital of Wuhan University。
文摘The dysbiosis of oral microbiota contributes to diseases such as periodontitis and certain cancers by triggering the host inflammatory response.Developing methods for the immediate and sensitive identification of oral microorganism is crucial for the rapid diagnosis and early interventions of associated diseases.Traditional methods for microbial detection primarily include the plate culturing,polymerase chain reaction and enzyme-linked immunosorbent assay,which are either time-consuming or laborious.Herein,we reported a persistent luminescence-encoded multiple-channel optical sensing array and achieved the rapid and accurate identification of oral-derived microorganisms.Our results demonstrate that electrostatic attractions and hydrophobic-hydrophobic interactions dominate the binding of the persistent luminescent nanoprobes to oral microorganisms and the microbial identification process can be finished within 30 min.Specifically,a total of 7 oral-derived microorganisms demonstrate their own response patterns and were differentiated by linear discriminant analysis(LDA)with the accuracy up to 100%both in the solution and artificial saliva samples.Moreover,the persistent luminescence encoded array sensor could also discern the microorganism mixtures with the accuracy up to 100%.The proposed persistent luminescence encoding sensor arrays in this work might offer new ideas for rapid and accurate oralderived microorganism detection,and provide new ways for disease diagnosis associated with microbial metabolism.
基金supported by the Innovation Capability Improvement Project of Hebei Province,China(Grant No.22567605H).
文摘We study the existence and stability of dark-gap solitons in linear lattice and nonlinear lattices.The results indicate that the combination of linear and nonlinear lattices gives dark-gap solitons unique properties.The linear lattice can stabilize dark-gap solitons,while the nonlinear lattice reduces the stability of dark-gap solitons.On the basis of numerical analysis,we investigate the effects of lattice depth,chemical potential,nonlinear lattice amplitude,and nonlinear lattice period on the soliton in mixed lattices with the same and different periods.The stability of dark-gap soliton is studied carefully by means of real-time evolution and linear stability analysis.Dark-gap solitons can exist stably in the band gap,but the solitons formed by the mixed lattices are slightly different when the period is the same or different.
基金supported by the National Natural Science Foundation of China(No. 50478075)the Jiangsu "Six Top Talent" Program of China(No. 07-F-008)+1 种基金the Priority Academic Program Development of Jiangsu Higher Education Institutionsthe Scientific Research Foundation of Graduate School of Southeast University (No. YBJJ0817),China
文摘Alternate path(AP) method is the most widely used method for the progressive collapse analysis,and its application in frame structures has been well proved.However,the application of AP method for other structures,especially for cable-stayed structures,should be further developed.The four analytical procedures,i.e.,linear static,nonlinear static,linear dynamic,and nonlinear dynamic were firstly improved by taking into account the initial state.Then a cable-stayed structure was studied using the four improved methods.Furthermore,the losses of both one cable and two cables were discussed.The results show that for static and dynamic analyses of the cable-stayed bridges,there is large difference between the results obtained from simulations starting with either a deformed or a nondeformed configuration at the time of cable loss.The static results are conservative in the vicinity of the ruptured cable,but the dynamic effect of the cable loss in the area farther away from the loss-cable cannot be considered.Moreover,the dynamic amplification factor of 2.0 is found to be a good estimate for static analysis procedures,since linear static and linear dynamic procedures yield approximately the same maximum vertical deflection.The results of the comprehensive evaluation of the cable failure show that the tread of the progressive failure of the cable-stayed bridges decreases when the location of the failed cables is closer to the pylon.
基金The National Natural Science Foundation of China (No.61374194)
文摘A direct linear discriminant analysis algorithm based on economic singular value decomposition （DLDA/ESVD） is proposed to address the computationally complex problem of the conventional DLDA algorithm, which directly uses ESVD to reduce dimension and extract eigenvectors corresponding to nonzero eigenvalues. Then a DLDA algorithm based on column pivoting orthogonal triangular （QR） decomposition and ESVD （DLDA/QR-ESVD） is proposed to improve the performance of the DLDA/ESVD algorithm by processing a high-dimensional low rank matrix, which uses column pivoting QR decomposition to reduce dimension and ESVD to extract eigenvectors corresponding to nonzero eigenvalues. The experimental results on ORL, FERET and YALE face databases show that the proposed two algorithms can achieve almost the same performance and outperform the conventional DLDA algorithm in terms of computational complexity and training time. In addition, the experimental results on random data matrices show that the DLDA/QR-ESVD algorithm achieves better performance than the DLDA/ESVD algorithm by processing high-dimensional low rank matrices.
基金co-supported by the National Basic Research Program of China(No.2014CB744802)Major Research of National Natural Science Foundation of China(No.91952302)China Postdoctoral Science Foundation(No.2018 M642007)。
文摘The Stereo Particle Image Velocimetry(SPIV)technology is applied to measure the wingtip vortices generated by the up-down symmetrical split winglet.Then,the temporal biglobal Linear Stability Analysis(bi-global LSA)is performed on this nearly equal-strength corotating vortex pair,which is composed of an upper vortex(vortex-u)and a down vortex(vortex-d).The results show that the instability eigenvalue spectrum illustrated by(ωr,ω_(i))contains two types of branches:discrete branch and continuous branch.The discrete branch contains the primary branches of vortex-u and vortex-d,the secondary branch of vortex-d and coupled branch,of which all of the eigenvalues are located in the unstable half-plane ofω_(i)>0,indicating that the wingtip vortex pair is temporally unstable.By contrast,the eigenvalues of the continuous branch are concentrated on the half-plane ofω_(i)<0 and the perturbation modes correspond to the freestream perturbation.In the primary branches of vortex-u and vortex-d,Mode P_(u) and Mode Pd are the primary perturbation modes,which exhibit the structures enclosed with azimuthal wavenumber m and radial wavenumber n,respectively.Besides,the results of stability curves for vortex-u and vortex-d demonstrate that the instability growth rates of vortex-u are larger than those of vortex-d,and the perturbation energy of Mode P_(u) is also larger than that of Mode Pd.Moreover,the perturbation energy of Mode P_(u) is up to 0.02650 and accounts for 33.56%percent in the corresponding branch,thereby indicating that the instability development of wingtip vortex is dominated by Mode P_(u).By further investigating the topological structures of Mode P_(u) and Mode Pd with streamwise wavenumbers,the most unstable perturbation mode with a large azimuthal wavenumber of m=5-6 is identified,which imposes on the entire core region of vortex-u.This large azimuthal wavenumber perturbation mode can suggest the potential physical-based flow control strategy by manipulating it.
基金supported by the National Natural Science Foundation of China(No.40572081)the Chinese Academy of Sciences(No.KZCX1-SW-18)
文摘Four samples from a Permian reservoir in the Ordos Basin of North China were separated into twelve fractions in grain sizes of 〈0.5, 0.5-1 and 1-2 μm. Using the ^40Ar-^39Ar step-heating method, all of the fractions essentially yielded plateau ages ranging from 172.5 to 217.1 Ma. These scattered plateau ages might not have been obtained from pure diagenetic illites but from mixed clay minerals, although the samples were disaggregated using a gentle freeze-thaw cycle to free them of non-clay minerals. A regional thermal event, as suggested by several proxies, led to intensive iHitization as a distinct diagenetic process when the Yanshanian Movement triggered magmatism around the entire North China Block during the Jurassic to Cretaceous. Thermal illites formed during a short time period, whereas detrital illites came from various sources. The scattered plateau ages could have resulted from mixed degassing of thermal and detrital illites. Within one sample, the plateau ages decrease with the diminution of grain sizes, but it is difficult to extrapolate to the detrital-illite-free ages. Because the plateau age is a mixture of ages for thermal and detrital illites, this regression analysis studies the dependence of the plateau ages on the synthetic values of contents and ages of detrital illites instead of on the grain sizes. Comparing the samples to one another, the plateau ages show the same trend among the different grain sizes. Weighted by the contents and ages of detrital illites, linear regression analysis revealed the relationship between the plateau ages and the relative weight proportions. Based on iterated calculations, a thermal event age and a set of weight proportions were derived. The regressed thermal event age is 163.3±1.6 Ma, which coincides with regional thermal activities and links to gas accumulation.
基金the National Natural Science Foundation of China (No. 50575145)the National High Technology Research and Development Program (863) of China(Nos. 2006AA04Z432 and 2007AA04Z419)
文摘Vehicle traveling time prediction is an important part of the research of intelligent transportation system. By now, there have been various kinds of methods for vehicle traveling time prediction. But few consider both aspects of time and space. In this paper, a vehicle traveling time prediction method based on grey theory （GT） and linear regression analysis （LRA） is presented. In aspects of time, we use the history data sequence of bus speed on a certain road to predict the future bus speed on that road by GT. And in aspects of space, we calculate the traffic affecting factors between various roads by LRA. Using these factors we can predict the vehicle＇s speed at the lower road if the vehicle＇s speed at the current road is known. Finally we use time factor and space factor as the weighting factors of the two results predicted by GT and LRA respectively to find the final result, thus calculating the vehicle＇s traveling time. The method also considers such factors as dwell time, thus making the prediction more accurate.
基金acknowledge thorough support from the Global COE program from the Ministry of Education, Culture, Sports and Technology (MEXT) of Japan
文摘A series of linear stability analysis is carried out on the onset of thermal convection in the presence of spatial variations of viscosity, thermal conductivity and expansivity. We consider the temporal evolution of an infinitesimal perturbation superimposed to a static （motionless） and con- ductive state in a basally-heated planar layer. From the changes in flow patterns with increasing the amplitudes of temperature dependence of viscosity, we identified the transition into the ＂stagnant-lid＂ （ST） regime, where the convection occurs only beneath a thick and stagnant-lid of cold fluid at the top surface. Detailed analysis showed a significant increase of the aspect ratio of convection cells in ST regime induced by the spatial variations in thermal conductivity and/or expansivity： the horizon- tal length scale of ST convection can be enlarged by up to 50% with 10 times increase of thermal conductivity with depth. We further developed an analytical model of ST convection which success- fully reproduced the mechanism of increasing horizontal length scale of ST regime convection cells for given spatial variations in physical properties. Our findings may highlight the essential roles of the spatial variation of thermal conductivity on the convection patterns in the mantle.
基金This project was funded by the Deanship of Scientific Research(DSR),King Abdulaziz University,Jeddah,Saudi Arabia under Grant No.(KEP-81-130-42)The authors,therefore acknowledge with thanks DSR technical and financial support。
文摘Security is a vital parameter to conserve energy in wireless sensor networks(WSN).Trust management in the WSN is a crucial process as trust is utilized when collaboration is important for accomplishing trustworthy data transmission.But the available routing techniques do not involve security in the design of routing techniques.This study develops a novel statistical analysis with dingo optimizer enabled reliable routing scheme(SADO-RRS)for WSN.The proposed SADO-RRS technique aims to detect the existence of attacks and optimal routes in WSN.In addition,the presented SADORRS technique derives a new statistics based linear discriminant analysis(LDA)for attack detection,Moreover,a trust based dingo optimizer(TBDO)algorithm is applied for optimal route selection in the WSN and accomplishes secure data transmission in WSN.Besides,the TBDO algorithm involves the derivation of the fitness function involving different input variables of WSN.For demonstrating the enhanced outcomes of the SADO-RRS technique,a wide range of simulations was carried out and the outcomes demonstrated the enhanced outcomes of the SADO-RRS technique.
基金support of National Key Research and Development Program of China(2020YFA0908303)National Natural Science Foundation of China(21878081).
文摘Visual process monitoring is important in complex chemical processes.To address the high state separation of industrial data,we propose a new criterion for feature extraction called balanced multiple weighted linear discriminant analysis(BMWLDA).Then,we combine BMWLDA with self-organizing map(SOM)for visual monitoring of industrial operation processes.BMWLDA can extract the discriminative feature vectors from the original industrial data and maximally separate industrial operation states in the space spanned by these discriminative feature vectors.When the discriminative feature vectors are used as the input to SOM,the training result of SOM can differentiate industrial operation states clearly.This function improves the performance of visual monitoring.Continuous stirred tank reactor is used to verify that the class separation performance of BMWLDA is more effective than that of traditional linear discriminant analysis,approximate pairwise accuracy criterion,max–min distance analysis,maximum margin criterion,and local Fisher discriminant analysis.In addition,the method that combines BMWLDA with SOM can effectively perform visual process monitoring in real time.
基金supported by the Science and Technology Project of Guangdong Province of China(Nos.2014A020213016 and 2014A020212445).
文摘The moving window bis corelation coefficients(MW BiCC)was proposed and employed for the discriminant analysis of transgenic sugarcane leaves and B-thalassemia with visible and near-infrared(Vis NIR)spectroscopy.The well-performed moving window principal component analysis linear discriminant analysis(MWPCA-LDA)was also conducted for comparison.A total of 306 transgenic(positive)and 150 nont ransgenic(negative)leave samples of sugarcane were collected and divided to calibration,prediction,and validation.The diffuse reflection spectra were corected using Savitzky-Golay(SG)smoothing with first-order derivative(d=1),third-degree polynomial(p=3)and 25 smpothing points(m=25).The selected waveband was 736-1054nm with MW-BiCC,and the positive and negative validation recognition rates(V_REC^(+),VREC^(-))were 100%,98.0%,which achieved the same effect as MWPCA-LDA.Another example,the 93 B-thalassemia(positive)and 148 nonthalassemia(negative)of human hemolytic samples were colloctod.The transmission spectra were corrected using SG smoothing withd=1,p=3 and m=53.Using M W-BiCC,many best wavebands were selected(e.g.,1116-1146,17941848 and 22842342nm).The V_REC^(+)and V_REC^(-)were both 100%,which achieved the same effect as MW-PCA-LDA.Importantly,the BICC only required ca lculating correlation cofficients between the spectrum of prediction sample and the average spectra of two types of calibration samples.Thus,BiCC was very simple in algorithm,and expected to obtain more applications.The results first confirmed the feasibility of distinguishing B-thalassemia and normal control samples by NIR spectroscopy,and provided a promising simple tool for large population thalassemia screening.
文摘This study proposes an approach based on machine learning to forecast currency exchange rates by applying sentiment analysis to messages on Twitter(called tweets).A dataset of the exchange rates between the United States Dollar(USD)and the Pakistani Rupee(PKR)was formed by collecting information from a forex website as well as a collection of tweets from the business community in Pakistan containing finance-related words.The dataset was collected in raw form,and was subjected to natural language processing by way of data preprocessing.Response variable labeling was then applied to the standardized dataset,where the response variables were divided into two classes:“1”indicated an increase in the exchange rate and“−1”indicated a decrease in it.To better represent the dataset,we used linear discriminant analysis and principal component analysis to visualize the data in three-dimensional vector space.Clusters that were obtained using a sampling approach were then used for data optimization.Five machine learning classifiers—the simple logistic classifier,the random forest,bagging,naïve Bayes,and the support vector machine—were applied to the optimized dataset.The results show that the simple logistic classifier yielded the highest accuracy of 82.14%for the USD and the PKR exchange rates forecasting.
基金funded by the National Basic Research Program of China (2013CB956000)the National Natural Science Foundation of China (41171010,41371102,41301003)
文摘In terms of formation mechanisms of linear dunes,there are open arguments for their widespread distribution and multi-morphological diversities.In order to clarify the formation mechanism of linear dunes of Qarhan Salt Lake,we used pattern analysis method to analyze the statistical characteristics and spatial variation of their pattern parameters.Except at the west-northwest margin,the pattern parameters showed regular spatial variation from the up-middle part towards the downwind end of the dune field.Based on the cumulative probability plots for inter-crest spacing and crest length,we divided the linear dunes into three groups,which corresponding to the three evolution stages of these dunes.The first group comprises erosional relics,with shorter crests,smaller inter-crest spacing and more random dune orientation.The second group comprises dunes whose sand supply is just sufficient to maintain stability and these dunes are approaching the net erosion stage.The crest length and inter-crest spacing of these dunes are much larger than those of the first group,and dune orientation is closer to the resultant drift direction （RDD） .The last group comprises linear dunes that are still undergoing vertical accretion and longitudinal elongation,which follows the RDD of the modern wind regime.The presence of regular spatial variation of pattern parameters and a similar geometry with the vegetated linear dunes suggest that deposition and erosion coexist in the development and evolution of linear dunes of Qarhan Salt Lake,i.e.deposition predominates at the downwind end of linear dunes in the vertical accretion and longitudinal elongation stage,whereas erosion mainly occurs at the upwind end of linear dunes in the degradation stage.Therefore,the formation mechanism of linear dunes in Qarhan Salt Lake can be reasonably explained by the combination of depositional and erosional theories.
文摘A kernel-based discriminant analysis method called kernel direct discriminant analysis is employed, which combines the merit of direct linear discriminant analysis with that of kernel trick. In order to demonstrate its better robustness to the complex and nonlinear variations of real face images, such as illumination, facial expression, scale and pose variations, experiments are carried out on the Olivetti Research Laboratory, Yale and self-built face databases. The results indicate that in contrast to kernel principal component analysis and kernel linear discriminant analysis, the method can achieve lower （7%） error rate using only a very small set of features. Furthermore, a new corrected kernel model is proposed to improve the recognition performance. Experimental results confirm its superiority （1% in terms of recognition rate） to other polynomial kernel models.
文摘Linear discrimiant analysis (LDA) has been used in face recognition. But it is difficult to handle the high nonlinear problems, such as changes of large viewpoint and illumination. In order to overcome these problems, kernel discriminant analysis for face recognition is presented. This approach adopts the kernel functions to replace the dot products of nonlinear mapping in the high dimensional feature space, and then the nonlinear problem can be solved in the input space conveniently without explicit mapping. Two face databases are given.
