Emotional speaker recognition based on prosody transformation

A novel emotional speaker recognition system （ESRS） is proposed to compensate for emotion variability. First, the emotion recognition is adopted as a pre-processing part to classify the neutral and emotional speech. Then, the recognized emotion speech is adjusted by prosody modification. Different methods including Gaussian normalization, the Gaussian mixture model （GMM） and support vector regression （SVR） are adopted to define the mapping rules of F0s between emotional and neutral speech, and the average linear ratio is used for the duration modification. Finally, the modified emotional speech is employed for the speaker recognition. The experimental results show that the proposed ESRS can significantly improve the performance of emotional speaker recognition, and the identification rate （IR） is higher than that of the traditional recognition system. The emotional speech with F0 and duration modifications is closer to the neutral one.

Modified algorithm of principal component analysis for face recognition

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.

Enhanced kernel minimum squared error algorithm and its application in face recognition

To improve the classification performance of the kernel minimum squared error（ KMSE）, an enhanced KMSE algorithm（ EKMSE） is proposed. It redefines the regular objective function by introducing a novel class label definition, and the relative class label matrix can be adaptively adjusted to the kernel matrix.Compared with the common methods, the newobjective function can enlarge the distance between different classes, which therefore yields better recognition rates. In addition, an iteration parameter searching technique is adopted to improve the computational efficiency. The extensive experiments on FERET and GT face databases illustrate the feasibility and efficiency of the proposed EKMSE. It outperforms the original MSE, KMSE,some KMSE improvement methods, and even the sparse representation-based techniques in face recognition, such as collaborate representation classification（ CRC）.

Feature fusing in face recognition

With the aim of extracting the features of face images in face recognition, a new method of face recognition by fusing global features and local features is presented. The global features are extracted using principal component analysis （PCA）. Active appearance model （AAM） locates 58 facial fiducial points, from which 17 points are characterized as local features using the Gabor wavelet transform （GWT）. Normalized global match degree （local match degree） can be obtained by global features （local features） of the probe image and each gallery image. After the fusion of normalized global match degree and normalized local match degree, the recognition result is the class that included the gallery image corresponding to the largest fused match degree. The method is evaluated by the recognition rates over two face image databases （AR and SJTU-IPPR）. The experimental results show that the method outperforms PCA and elastic bunch graph matching （EBGM）. Moreover, it is effective and robust to expression, illumination and pose variation in some degree.

Real-Time Face Tracking and Recognition in Video Sequence

A framework of real time face tracking and recognition is presented, which integrates skin color based tracking and PCA/BPNN (principle component analysis/back propagation neural network) hybrid recognition techniques. The algorithm is able to track the human face against a complex background and also works well when temporary occlusion occurs. We also obtain a very high recognition rate by averaging a number of samples over a long image sequence. The proposed approach has been successfully tested by many experiments, and can operate at 20 frames/s on an 800 MHz PC.

Comparative Study on VQ-Based Efficient Mandarin Speech Recognition Method

A VQ based efficient speech recognition method is introduced, and the key parameters of this method are comparatively studied. This method is especially designed for mandarin speaker dependent small size word set recognition. It has less complexity, less resource consumption but higher ARR (accurate recognition rate) compared with traditional HMM or NN approach. A large scale test on the task of 11 mandarin digits recognition shows that the WER(word error rate) can reach 3 86%. This method is suitable for being embedded in PDA (personal digital assistant), mobile phone and so on to perform voice controlling like digits dialing, name dialing, calculating, voice commanding, etc.

LOCAL BAGGING AND ITS APPLICATIONON FACE RECOGNITION

Bagging is not quite suitable for stable classifiers such as nearest neighbor classifiers due to the lack of diversity and it is difficult to be directly applied to face recognition as well due to the small sample size （SSS） property of face recognition. To solve the two problems,local Bagging （L-Bagging） is proposed to simultaneously make Bagging apply to both nearest neighbor classifiers and face recognition. The major difference between L-Bagging and Bagging is that L-Bagging performs the bootstrap sampling on each local region partitioned from the original face image rather than the whole face image. Since the dimensionality of local region is usually far less than the number of samples and the component classifiers are constructed just in different local regions,L-Bagging deals with SSS problem and generates more diverse component classifiers. Experimental results on four standard face image databases （AR,Yale,ORL and Yale B） indicate that the proposed L-Bagging method is effective and robust to illumination,occlusion and slight pose variation.

FUZZY WITHIN-CLASS MATRIX PRINCIPAL COMPONENT ANALYSIS AND ITS APPLICATION TO FACE RECOGNITION

Matrix principal component analysis （MatPCA）, as an effective feature extraction method, can deal with the matrix pattern and the vector pattern. However, like PCA, MatPCA does not use the class information of samples. As a result, the extracted features cannot provide enough useful information for distinguishing pat- tern from one another, and further resulting in degradation of classification performance. To fullly use class in- formation of samples, a novel method, called the fuzzy within-class MatPCA （F-WMatPCA）is proposed. F-WMatPCA utilizes the fuzzy K-nearest neighbor method（FKNN） to fuzzify the class membership degrees of a training sample and then performs fuzzy MatPCA within these patterns having the same class label. Due to more class information is used in feature extraction, F-WMatPCA can intuitively improve the classification perfor- mance. Experimental results in face databases and some benchmark datasets show that F-WMatPCA is effective and competitive than MatPCA. The experimental analysis on face image databases indicates that F-WMatPCA im- proves the recognition accuracy and is more stable and robust in performing classification than the existing method of fuzzy-based F-Fisherfaces.

A novel speech emotion recognition algorithm based on combination of emotion data field and ant colony search strategy

In order to effectively conduct emotion recognition from spontaneous, non-prototypical and unsegmented speech so as to create a more natural human-machine interaction; a novel speech emotion recognition algorithm based on the combination of the emotional data field （EDF） and the ant colony search （ACS） strategy, called the EDF-ACS algorithm, is proposed. More specifically, the inter- relationship among the turn-based acoustic feature vectors of different labels are established by using the potential function in the EDF. To perform the spontaneous speech emotion recognition, the artificial colony is used to mimic the turn- based acoustic feature vectors. Then, the canonical ACS strategy is used to investigate the movement direction of each artificial ant in the EDF, which is regarded as the emotional label of the corresponding turn-based acoustic feature vector. The proposed EDF-ACS algorithm is evaluated on the continueous audio）＇visual emotion challenge （AVEC） 2012 dataset, which contains the spontaneous, non-prototypical and unsegmented speech emotion data. The experimental results show that the proposed EDF-ACS algorithm outperforms the existing state-of-the-art algorithm in turn-based speech emotion recognition.

A Study of the Gendered Identity of Female Migrants in Contemporary China

As a new group in the process of socio-economic transformation, female members of the Chinese floating population have negative self-identities rooted in the traditional gender discrimination in China society. This impacts negatively on female migrants1 both physiologically and psychologically, disadvantaging them in the pursuit of resources, opportunities and rights. It is therefore necessary to positively influence the self-image of female migrants so as to ensure their rights, further their interests and ultimately to achieve gender equality.

DUAL-SPARSITY PRESERVING PROJECTION

Sparsity preserving projection（SPP） is a popular graph-based dimensionality reduction（DR） method, which has been successfully applied to solve face recognition recently. SPP contains natural discriminating information by preserving sparse reconstruction relationship of data sets. However, SPP suffers from the fact that every new feature learned from data sets is linear combinations of all the original features, which often makes it difficult to interpret the results. To address this issue, a novel DR method called dual-sparsity preserving projection （DSPP） is proposed to further impose sparsity constraints on the projection directions of SPP. Specifically, the proposed method casts the projection function learning of SPP into a regression-type optimization problem, and then the sparse projections can be efficiently computed by the related lasso algorithm. Experimental results from face databases demonstrate the effectiveness of the proposed algorithm.

A new reaction system to determine nonlinear chemical fingerprint and its use in Panax ginseng identification method based on double reaction system

A new reaction system to determine nonlinear chemical fingerprint(NCF)and its use in identification method based on double reaction system was researched.Panax ginsengs,such as ginseng,American ginseng and notoginseng were identified by the method.The NCFs of the three samples of Panax ginsengs were determined through two nonlinear chemical systems,namely system 1 consisting of sample components,H2SO4,MnSO4,NaBrO3,acetone and the new system,system 2 consisting of sample components,H2SO4,(NH4)4Ce(SO4)2,NaBrO3 and citric acid.The comparison between the results determined through systems 1 and 2 shows that the speed to determine NCF through system 2 is much faster than that through system 1;for systems 1 and 2,the system similarities of the same kind of samples are≥98.09%and 99.78%,respectively,while those of different kinds of samples are≤63.04%and 86.34%,respectively.The results to identify the kinds of some samples by system similarity pattern show that both the accuracies of identification methods based on single system 1 and 2 are≥95.6%,and the average values are 97.1%and 96.3%,respectively;the accuracy of the method based on double system is≥97.8%,and the average accuracy is 99.3%.The accuracy of the method based on double system is higher than that based on any single system.

Kernel principal component analysis network for image classification

In order to classify nonlinear features with a linear classifier and improve the classification accuracy, a deep learning network named kernel principal component analysis network（ KPCANet） is proposed. First, the data is mapped into a higher-dimensional space with kernel principal component analysis to make the data linearly separable. Then a two-layer KPCANet is built to obtain the principal components of the image. Finally, the principal components are classified with a linear classifier. Experimental results showthat the proposed KPCANet is effective in face recognition, object recognition and handwritten digit recognition. It also outperforms principal component analysis network（ PCANet） generally. Besides, KPCANet is invariant to illumination and stable to occlusion and slight deformation.

Hand Gesture Recognition Based on Improved FRNN

The trained Gaussian mixture model is used to make skincolour segmentation for the input image sequences. The hand gesture region is extracted, and the relative normalization images are obtained by interpolation operation. To solve the proem of hand gesture recognition, Fuzzy-Rough based nearest neighbour（RNN） algorithm is applied for classification. For avoiding the costly compute, an improved nearest neighbour classification algorithm based on fuzzy-rough set theory （FRNNC） is proposed. The algorithm employs the represented cluster points instead of the whole training samples, and takes the hand gesture data＇s fuzziness and the roughness into account, so the campute spending is decreased and the recognition rate is increased. The 30 gestures in Chinese sign language alphabet are used for approving the effectiveness of the proposed algorithm. The recognition rate is 94.96%, which is better than that of KNN （K nearest neighbor）and Fuzzy- KNN （Fuzzy K nearest neighbor）.

Active Appearance Model Based Hand Gesture Recognition

This paper addresses the application of hand gesture recognition in monocular image sequences using Active Appearance Model （AAM）, For this work, the proposed algorithm is composed of constricting AAMs and fitting the models to the interest region. In training stage, according to the manual labeled feature points, the relative AAM is constructed and the corresponding average feature is obtained. In recognition stage, the interesting hand gesture region is firstly segmented by skin and movement cues. Secondly, the models are fitted to the image that includes the hand gesture, and the relative features are extracted. Thirdly, the classification is done by comparing the extracted features and average features. 30 different gestures of Chinese sign language are applied for testing the effectiveness of the method. The Experimental results are given indicating good performance of the algorithm.

Framework of Pattern Recognition Model Based on the Cognitive Psychology

According to the fundamental theory of visual cognition mechanism and cognitive psychology,the visual pattern recognition model is introduced briefly.Three pattern recognition models,i.e.template_based matching model,prototype_based matching model and feature_based matching model are built and discussed separately.In addition,the influence of object background information and visual focus point to the result of pattern recognition is also discussed with the example of recognition for fuzzy letters and figures

Detection and Classification on Amateur Drones Based on Cepstrum of Radio Frequency Signal

As a prospective component of the future air transportation system,unmanned aerial vehicles(UAVs)have attracted enormous interest in both academia and industry.However,small UAVs are barely supervised in the current situation.Crash accidents or illegal airspace invading caused by these small drones affect public security negatively.To solve this security problem,we use the back-propagation neural network(BPNN),the support-vector machine(SVM),and the k-nearest neighbors(KNN)method to detect and classify the non-cooperative drones at the edge of the flight restriction zone based on the cepstrum of the radio frequency(RF)signal of the drone’s downlink.The signal from five various amateur drones and ambient wireless devices are sampled in an electromagnetic clean environment.The detection and classification algorithm based on the cepstrum properties is conducted.Results of the outdoor experiments suggest the proposed workflow and methods are sufficient to detect non-cooperative drones with an average accuracy of around 90%.The mainstream downlink protocols of amateur drones can be classified effectively as well.

Identification of individuals with non-alcoholic fatty liver disease by the diagnostic criteria for the metabolic syndrome

AIM:To clarify the efficiency of the criterion of metabolic syndrome to detecting non-alcoholic fatty liver disease(NAFLD).METHODS:Authors performed a cross-sectional study involving participants of a medical health checkup program including abdominal ultrasonography.This study involved 11 714 apparently healthy Japanese men and women,18 to 83 years of age.NAFLD was defined by abdominal ultrasonography without an alcohol intake of more than 20 g/d,known liver disease,or current use of medication.The revised criteria of the National Cholesterol Education Program Adult Treatment PanelⅢ were used to characterize the metabolic syndrome.RESULTS:NAFLD was detected in 32.2%(95%CI:31.0%-33.5%)of men(n=1874 of 5811)and in 8.7%(95%CI:8.0%-9.5%)of women(n=514 of 5903).Among obese people,the prevalence of NAFLD was as high as 67.3%(95%CI:64.8%-69.7%)in men and 45.8%(95%CI:41.7%-50.0%)in women.Although NAFLD was thought of as being the liver phenotype of metabolic syndrome,the prevalence of the metabolic syndrome among subjects with NAFLD was low both in men and women.66.8%of men and 70.4%of women with NAFLD were not diagnosed with the metabolic syndrome.48.2%of men with NAFLD and 49.8%of women with NAFLD weren't overweight[body mass index(BMI)≥25 kg/m2].In the same way,68.6%of men with NAFLD and 37.9%of women with NAFLD weren't satisfied with abdominal classification(≥90 cm for men and≥80 cm for women).Next,authors defined it as positive at screening for NAFLD when participants satisfied at least one criterion of metabolic syndrome.The sensitivity of the definition"at least 1 criterion"was as good as 84.8%in men and 86.6%in women.Separating subjects by BMI,the sensitivity was higher in obese men and women than in non-obese men and women(92.3%vs 76.8%in men,96.1%vs 77.0%in women,respectively).CONCLUSION:Authors could determine NAFLD effectively in epidemiological study by modifying the usage of the criteria for metabolic syndrome.

Face Orientation Normalization Using Eye Positions

Despite the fact that progress in face recognition algorithms over the last decades has been made, changing lighting conditions and different face orientation still remain as a challenging problem. A standard face recognition system identifies the person by comparing the input picture against pictures of all faces in a database and finding the best match. Usually face matching is carried out in two steps： during the first step detection of a face is done by finding exact position of it in a complex background （various lightning condition）, and in the second step face identification is performed using gathered databases. In reality detected faces can appear in different position and they can be rotated, so these disturbances reduce quality of the recognition algorithms dramatically. In this paper to increase the identification accuracy we propose original geometric normalization of the face, based on extracted facial feature position such as eyes. For the eyes localization lbllowing methods has been used： color based method, mean eye template and SVM （Support Vector Machine） technique. Experimental investigation has shown that the best results for eye center detection can be achieved using SVM technique. The recognition rate increases statistically by 28% using face orientation normalization based on the eyes position.

Online RGB-D person re-identification based on metric model update

Person re-identification （re-id） on robot platform is an important application for human-robot- interaction （HRI）, which aims at making the robot recognize the around persons in varying scenes. Although many effective methods have been proposed for surveillance re-id in recent years, re-id on robot platform is still a novel unsolved problem. Most existing methods adapt the supervised metric learning offline to improve the accuracy. However, these methods can not adapt to unknown scenes. To solve this problem, an online re-id framework is proposed. Considering that robotics can afford to use high-resolution RGB-D sensors and clear human face may be captured, face information is used to update the metric model. Firstly, the metric model is pre-trained offline using labeled data. Then during the online stage, we use face information to mine incorrect body matching pairs which are collected to update the metric model online. In addition, to make full use of both appearance and skeleton information provided by RGB-D sensors, a novel feature funnel model （FFM） is proposed. Comparison studies show our approach is more effective and adaptable to varying environments.