An SVM-AdaBoost facial expression recognition system

This study is focused on improving the recognition rate and processing time of facial recognition systems. First, the skin is detected by pixel based methods to reduce the searching space for maximum rejection classifier (MRC) which detects the face. The detected face is normalized by a discrete cosine transform (DCT) and down-sampled by Bessel transform. Gabor feature extraction techniques were utilized to extract thousands of facial features that represent facial deformation patterns. An AdaBoost-based hypothesis is formulated to select a few hundreds of Gabor features which are potential candidates for expression recognition. The selected features were fed into a saturated vector machine (SVM) classifier to train it. An average recognition rate of 97.57 % and 92.33 % are registered in JAFFE and Yale databases respectively. The execution time of the proposed method is also significantly lower than others. Generally, the proposed method exhibits superior performance than other methods.     

A novel adaptive cuckoo search algorithm for intrinsic discriminant analysis based face recognition

This paper presents a novel adaptive cuckoo search (ACS) algorithm for optimization. The step size is made adaptive from the knowledge of its fitness function value and its current position in the search space. The other important feature of the ACS algorithm is its speed, which is faster than the CS algorithm. Here, an attempt is made to make the cuckoo search (CS) algorithm parameter free, without a Levy step. The proposed algorithm is validated using twenty three standard benchmark test functions. The second part of the paper proposes an efficient face recognition algorithm using ACS, principal component analysis (PCA) and intrinsic discriminant analysis (IDA). The proposed algorithms are named as PCA + IDA and ACS–IDA. Interestingly, PCA + IDA offers us a perturbation free algorithm for dimension reduction while ACS + IDA is used to find the optimal feature vectors for classification of the face images based on the IDA. For the performance analysis, we use three standard face databases—YALE, ORL, and FERET. A comparison of the proposed method with the state-of-the-art methods reveals the effectiveness of our algorithm.     

Genetic programming for epileptic pattern recognition in electroencephalographic signals

This paper reports how the genetic programming paradigm, in conjunction with pattern recognition principles, can be used to evolve classifiers capable of recognizing epileptic patterns in human electroencephalographic signals. The procedure for feature extraction from the raw signal is detailed, as well as the genetic programming system that properly selects the features and evolves the classifiers. Based on the data sets used, two different epileptic patterns were detected: 3 Hz spike-and-slow-wave-complex (SASWC) and spike-or-sharp-wave (SOSW). After training, classifiers for both patterns were tested with unseen instances, and achieved sensibility = 1.00 and specificity = 0.93 for SASWC patterns, and sensibility = 0.94 and specificity = 0.89 for SOSW patterns. Results are very promising and suggest that the methodology presented can be applied to other pattern recognition tasks in complex signals.     

A new approach of audio emotion recognition

A new architecture of intelligent audio emotion recognition is proposed in this paper. It fully utilizes both prosodic and spectral features in its design. It has two main paths in parallel and can recognize 6 emotions. Path 1 is designed based on intensive analysis of different prosodic features. Significant prosodic features are identified to differentiate emotions. Path 2 is designed based on research analysis on spectral features. Extraction of Mel-Frequency Cepstral Coefficient (MFCC) feature is then followed by Bi-directional Principle Component Analysis (BDPCA), Linear Discriminant Analysis (LDA) and Radial Basis Function (RBF) neural classification. This path has 3 parallel BDPCA + LDA + RBF sub-paths structure and each handles two emotions. Fusion modules are also proposed for weights assignment and decision making. The performance of the proposed architecture is evaluated on eNTERFACE’05 and RML databases. Simulation results and comparison have revealed good performance of the proposed recognizer.     

Action recognition via bio-inspired features: The richness of center–surround interaction

Motion is a key feature for a wide class of computer vision approaches to recognize actions. In this article, we show how to define bio-inspired features for action recognition. To do so, we start from a well-established bio-inspired motion model of cortical areas V1 and MT. The primary visual cortex, designated as V1, is the first cortical area encountered in the visual stream processing and early responses of V1 cells consist in tiled sets of selective spatiotemporal filters. The second cortical area of interest in this article is area MT where MT cells pool incoming information from V1 according to the shape and characteristic of their receptive field. To go beyond the classical models and following the observations from Xiao et al. [61], we propose here to model different surround geometries for MT cells receptive fields. Then, we define the so-called bio-inspired features associated to an input video, based on the average activity of MT cells. Finally, we show how these features can be used in a standard classification method to perform action recognition. Results are given for the Weizmann and KTH databases. Interestingly, we show that the diversity of motion representation at the MT level (different surround geometries), is a major advantage for action recognition. On the Weizmann database, the inclusion of different MT surround geometries improved the recognition rate from 63.01 ± 2.07% up to 99.26 ± 1.66% in the best case. Similarly, on the KTH database, the recognition rate was significantly improved with the inclusion of MT different surround geometries (from 47.82 ± 2.71% up to 92.44 ± 0.01% in the best case). We also discussed the limitations of the current approach which are closely related to the input video duration. These promising results encourage us to further develop bio-inspired models incorporating other brain mechanisms and cortical areas in order to deal with more complex videos.     

基于压缩金字塔核稀疏表示的人脸识别

人脸识别中光照、伪装及姿态等变化一直是富有挑战性的问题,其中特征提取是很关键的一步。为提高人脸识别率,结合压缩感知和空间金字塔模型,本文提出了一种新的特征提取方法,首先用尺度不变特征变换算法提取图像特征,然后与随机生成的字典进行稀疏编码,再用金字塔模型分层提取不同尺度空间的特征,并用最大池融合特征,最后运用核稀疏表示分类。在Extended Yale B,AR 和CMU PIE人脸数据库上的实验结果表明,该方法对于人脸图像的光照、伪装及姿态等变化有较强的鲁棒性,而且该算法有较快的运行速度。     

Human activity recognition with smartphone sensors using deep learning neural networks

Human activities are inherently translation invariant and hierarchical. Human activity recognition (HAR), a field that has garnered a lot of attention in recent years due to its high demand in various application domains, makes use of time-series sensor data to infer activities. In this paper, a deep convolutional neural network (convnet) is proposed to perform efficient and effective HAR using smartphone sensors by exploiting the inherent characteristics of activities and 1D time-series signals, at the same time providing a way to automatically and data-adaptively extract robust features from raw data. Experiments show that convnets indeed derive relevant and more complex features with every additional layer, although difference of feature complexity level decreases with every additional layer. A wider time span of temporal local correlation can be exploited (1 × 9–1 × 14) and a low pooling size (1 × 2–1 × 3) is shown to be beneficial. Convnets also achieved an almost perfect classification on moving activities, especially very similar ones which were previously perceived to be very difficult to classify. Lastly, convnets outperform other state-of-the-art data mining techniques in HAR for the benchmark dataset collected from 30 volunteer subjects, achieving an overall performance of 94.79% on the test set with raw sensor data, and 95.75% with additional information of temporal fast Fourier transform of the HAR data set.     

Tuning and hybrid parallelization of a genetic-based multi-point statistics simulation code

One of the main difficulties using multi-point statistical (MPS) simulation based on annealing techniques or genetic algorithms concerns the excessive amount of time and memory that must be spent in order to achieve convergence. In this work we propose code optimizations and parallelization schemes over a genetic-based MPS code with the aim of speeding up the execution time. The code optimizations involve the reduction of cache misses in the array accesses, avoid branching instructions and increase the locality of the accessed data. The hybrid parallelization scheme involves a fine-grain parallelization of loops using a shared-memory programming model (OpenMP) and a coarse-grain distribution of load among several computational nodes using a distributed-memory programming model (MPI). Convergence, execution time and speed-up results are presented using 2D training images of sizes 100 × 100 × 1 and 1000 × 1000 × 1 on a distributed-shared memory supercomputing facility.     

Spectral entropy feature subset selection using SEPCOR to detect alcoholic impact on gamma sub band visual event related potentials of multichannel electroencephalograms (EEG)

The problem of analyzing and identifying regions of high discrimination between alcoholics and controls in a multichannel electroencephalogram (EEG) signal is modeled as a feature subset selection technique that can improve the recognition rate between both groups. Several studies have reported efficient detection of alcoholics by feature extraction and selection in gamma band visual event related potentials (ERP) of a multichannel EEG signal. However, in these studies the correlation between features and their class information is not considered for feature selection. This may lead to redundancy in the feature set and result in over fitting. Therefore in this study, a statistical feature selection technique based on Separability & Correlation analysis (SEPCOR) is proposed to select an optimal feature subset automatically that possesses minimum correlation between selected channels and maximum class separation. The optimal feature selection consists of a ranking method that assigns ranks to channels based on a variability measure (V-measure). From the ranked feature set of highly discriminative features, different subsets are automatically selected by heuristically applying a correlation threshold in steps from 0.02 to 0.1. These subsets are applied as input features to multilayer perceptron (MLP) neural network and k-nearest neighbor (k-NN) classifiers to discriminate alcoholic and control visual ERP. Prior to feature selection, spectral entropy features are computed in gamma sub band (30–55 Hz) interval of a 61-channel multi-trial EEG signal with multiple object recognition tasks. Independent Component Analysis (ICA) is performed on raw EEG data to remove eye blink, motion and muscle artifacts. Results indicate that both classifiers exhibit excellent classification accuracy of 99.6%, for a feature subset of 22 optimal channels with correlation threshold of 0.1. In terms of computation time, k-NN classifier outperforms multilayer perceptron-back propagation (MLP-BP) network with 7.93 s whereas MLP network takes 55 s to perform the recognition task with the same accuracy. Compared to feature section methods used in previous studies on the same EEG alcoholic database, there is a significant improvement in classification accuracy based on the proposed SEPCOR method.     

基于ENM-Gabor差分权重的人脸表情特征提取方法

针对人脸子区域对表情识别分类的重要程度不同,提出一种基于Gabor小波特征和ENM(Eye,Nose,Mouth)差分权重的表情特征提取方法。通过对人脸眼睛、鼻子、嘴巴三个区域进行特征提取并自适应加以权重,有效地区分了不同区域对识别表情的重要程度。对预处理后的表情图像提取ENM区域Gabor特征;将表情图像与中性图像作差值计算得到ENM差分权重;将ENM-Gabor特征结合差分权重得到最终的表情特征并用BP神经网络进行分类。与其他方法在JAFFE表情库上进行对比实验,实验结果表明,该方法相比于传统Gabor特征提取有了明显的提高,且平均识别率达到99.3%。     

Spider specie identification and verification based on pattern recognition of it cobweb

Biodiversity conservation is a global priority where the study of every type of living form is a fundamental task. Inside the huge number of the planet species, spiders play an important role in almost every habitat. This paper presents a comprehensive study on the reliability of the most used features extractors to face the problem of spider specie recognition by using their cobwebs, both in identification and verification modes. We have applied a preprocessing to the cobwebs images in order to obtain only the valid information and compute the optimal size to reach the highest performance. We have used the principal component analysis (PCA), independent component analysis (ICA), Discrete Cosine Transform (DCT), Wavelet Transform (DWT) and discriminative common vectors as features extractors, and proposed the fusion of several of them to improve the system’s performance. Finally, we have used the Least Square Vector Support Machine with radial basis function as a classifier. We have implemented K-Fold and Hold-Out cross-validation techniques in order to obtain reliable results. PCA provided the best performance, reaching a 99.65% ± 0.21 of success rate in identification mode and 99.98% ± 0.04 of the area under de Reveicer Operating Characteristic (ROC) curve in verification mode. The best combination of features extractors was PCA, DCT, DWT and ICA, which achieved a 99.96% ± 0.16 of success rate in identification mode and perfect verification.     

Evaluating the use of ECG signal in low frequencies as a biometry

Traditional strategies, such as fingerprinting and face recognition, are becoming more and more fraud susceptible. As a consequence, new and more fraud proof biometrics modalities have been considered, one of them being the heartbeat pattern acquired by an electrocardiogram (ECG). While methods for subject identification based on ECG signal work with signals sampled in high frequencies (>100 Hz), the main goal of this work is to evaluate the use of ECG signal in low frequencies for such aim. In this work, the ECG signal is sampled in low frequencies (30 Hz and 60 Hz) and represented by four feature extraction methods available in the literature, which are then feed to a Support Vector Machines (SVM) classifier to perform the identification. In addition, a classification approach based on majority voting using multiple samples per subject is employed and compared to the traditional classification based on the presentation of single samples per subject each time. Considering a database composed of 193 subjects, results show identification accuracies higher than 95% and near to optimality (i.e., 100%) when the ECG signal is sampled in 30 Hz and 60 Hz, respectively, being the last one very close to the ones obtained when the signal is sampled in 360 Hz (the maximum frequency existing in our database). We also evaluate the impact of: (1) the number of training and testing samples for learning and identification, respectively; (2) the scalability of the biometry (i.e., increment on the number of subjects); and (3) the use of multiple samples for person identification.     

基于张量分析的表情特征提取

表情识别的性能依赖于所提取表情特征的有效性,现有方法提取的表情基本上是人脸与表情的融合体,然而不同个体的人脸差异是表情识别的主要干扰因素。在表情识别时,理想情况是将个体相关的人脸特征和与个体无关的表情特征相分离。针对此问题,在三维空间建立人脸张量;然后用张量分析的方法将人脸特征与表情特征进行分离,使获取的表情参数与人脸无关。从而排除不同个体的人脸差异对表情识别的干扰。最后,在JAFFE表情数据库上验证了该方法的有效性。     

Unsupervised feature selection using swarm intelligence and consensus clustering for automatic fault detection and diagnosis in Heating Ventilation and Air Conditioning systems

Various sensory and control signals in a Heating Ventilation and Air Conditioning (HVAC) system are closely interrelated which give rise to severe redundancies between original signals. These redundancies may cripple the generalization capability of an automatic fault detection and diagnosis (AFDD) algorithm. This paper proposes an unsupervised feature selection approach and its application to AFDD in a HVAC system. Using Ensemble Rapid Centroid Estimation (ERCE), the important features are automatically selected from original measurements based on the relative entropy between the low- and high-frequency features. The materials used is the experimental HVAC fault data from the ASHRAE-1312-RP datasets containing a total of 49 days of various types of faults and corresponding severity. The features selected using ERCE (Median normalized mutual information (NMI) = 0.019) achieved the least redundancies compared to those selected using manual selection (Median NMI = 0.0199) Complete Linkage (Median NMI = 0.1305), Evidence Accumulation K-means (Median NMI = 0.04) and Weighted Evidence Accumulation K-means (Median NMI = 0.048). The effectiveness of the feature selection method is further investigated using two well-established time-sequence classification algorithms: (a) Nonlinear Auto-Regressive Neural Network with eXogenous inputs and distributed time delays (NARX-TDNN); and (b) Hidden Markov Models (HMM); where weighted average sensitivity and specificity of: (a) higher than 99% and 96% for NARX-TDNN; and (b) higher than 98% and 86% for HMM is observed. The proposed feature selection algorithm could potentially be applied to other model-based systems to improve the fault detection performance.     

An intelligent strategy for checking the annual inspection status of motorcycles based on license plate recognition

License plate recognition techniques have been successfully applied to the management of stolen cars, management of parking lots and traffic flow control. This study proposes a license plate based strategy for checking the annual inspection status of motorcycles from images taken along the roadside and at designated inspection stations. Both a UMPC (Ultra Mobile Personal Computer) with a web camera and a desktop PC are used as hardware platforms. The license plate locations in images are identified by means of integrated horizontal and vertical projections that are scanned using a search window. Moreover, a character recovery method is exploited to enhance the success rate. Character recognition is achieved using both a back propagation artificial neural network and feature matching. The identified license plate can then be compared with entries in a database to check the inspection status of the motorcycle. Experiments yield a recognition rate of 95.7% and 93.9% based on roadside and inspection station test images, respectively. It takes less than 1 s on a UMPC (Celeron 900 MHz with 256 MB memory) and about 293 ms on a PC (Intel Pentium 4 3.0 GHz with 1 GB memory) to correctly recognize a license plate. Challenges associated with recognizing license plates from roadside and designated inspection stations images are also discussed.     

Speech emotion recognition: Features and classification models

To solve the speaker independent emotion recognition problem, a three-level speech emotion recognition model is proposed to classify six speech emotions, including sadness, anger, surprise, fear, happiness and disgust from coarse to fine. For each level, appropriate features are selected from 288 candidates by using Fisher rate which is also regarded as input parameter for Support Vector Machine (SVM). In order to evaluate the proposed system, principal component analysis (PCA) for dimension reduction and artificial neural network (ANN) for classification are adopted to design four comparative experiments, including Fisher + SVM, PCA + SVM, Fisher + ANN, PCA + ANN. The experimental results proved that Fisher is better than PCA for dimension reduction, and SVM is more expansible than ANN for speaker independent speech emotion recognition. The average recognition rates for each level are 86.5%, 68.5% and 50.2% respectively.     

Guided local search as a network planning algorithm that incorporates uncertain traffic demands

A search based heuristic for the optimisation of communication networks where traffic forecasts are uncertain and the problem is NP-complete is presented. While algorithms such as genetic algorithms (GA) and simulated annealing (SA) are often used for this class of problem, this work applies a combination of newer optimisation techniques specifically: fast local search (FLS) as an improved hill climbing method and guided local search (GLS) to allow escape from local minima. The GLS + FLS combination is compared with an optimised GA and SA approaches. It is found that in terms of implementation, the parameterisation of the GLS + FLS technique is significantly simpler than that for a GA and SA. Also, the self-regularisation feature of the GLS + FLS approach provides a distinctive advantage over the other techniques which require manual parameterisation. To compare numerical performance, the three techniques were tested over a number of network sets varying in size, number of switch circuit demands (network bandwidth demands) and levels of uncertainties on the switch circuit demands. The results show that the GLS + FLS outperforms the GA and SA techniques in terms of both solution quality and optimisation speed but even more importantly GLS + FLS has significantly reduced parameterisation time.     

结合LBP特征和深度学习的人脸表情识别

对于人脸表情识别,传统方法是先提取图像特征,再使用机器学习方法进行识别,这种方法不但特征提取过程复杂且泛化能力也差。为了达到更好的人脸表情识别效果,文中提出一种结合特征提取和卷积神经网络的人脸表情识别方法。首先使用基于Haar-like特征的AdaBoost算法对于数据库原始图片进行人脸区域检测,然后提取人脸区域局部二值模式（Local Binary Patterns,LBP）特征图,将其尺寸归一化后输入到改进的LeNet-5神经网络模型中进行识别。在CK+和JAFFE数据集上采用10折交叉验证方法进行实验,分别为98.19%和96.35%的准确率。实验结果表明该方法与其他主流方法相比在人脸表情识别上有一定的先进性和有效性。