Application of higher order spectral features and support vector machines for bearing faults classification

Condition monitoring and fault diagnosis of rolling element bearings timely and accurately are very important to ensure the reliability of rotating machinery. This paper presents a novel pattern classification approach for bearings diagnostics, which combines the higher order spectra analysis features and support vector machine classifier. The use of non-linear features motivated by the higher order spectra has been reported to be a promising approach to analyze the non-linear and non-Gaussian characteristics of the mechanical vibration signals. The vibration bi-spectrum (third order spectrum) patterns are extracted as the feature vectors presenting different bearing faults. The extracted bi-spectrum features are subjected to principal component analysis for dimensionality reduction. These principal components were fed to support vector machine to distinguish four kinds of bearing faults covering different levels of severity for each fault type, which were measured in the experimental test bench running under different working conditions. In order to find the optimal parameters for the multi-class support vector machine model, a grid-search method in combination with 10-fold cross-validation has been used. Based on the correct classification of bearing patterns in the test set, in each fold the performance measures are computed. The average of these performance measures is computed to report the overall performance of the support vector machine classifier. In addition, in fault detection problems, the performance of a detection algorithm usually depends on the trade-off between robustness and sensitivity. The sensitivity and robustness of the proposed method are explored by running a series of experiments. A receiver operating characteristic (ROC) curve made the results more convincing. The results indicated that the proposed method can reliably identify different fault patterns of rolling element bearings based on vibration signals.     

Facial expression recognition of intercepted video sequences based on feature point movement trend and feature block texture variation

Facial Expression Recognition (FER) is an important subject of human–computer interaction and has long been a research area of great interest. Accurate Facial Expression Sequence Interception (FESI) and discriminative expression feature extraction are two enormous challenges for the video-based FER. This paper proposes a framework of FER for the intercepted video sequences by using feature point movement trend and feature block texture variation. Firstly, the feature points are marked by Active Appearance Model (AAM) and the most representative 24 of them are selected. Secondly, facial expression sequence is intercepted from the face video by determining two key frames whose emotional intensities are minimum and maximum, respectively. Thirdly, the trend curve which represents the Euclidean distance variations between any two selected feature points is fitted, and the slopes of specific points on the trend curve are calculated. Finally, combining Slope Set which is composed by the calculated slopes with the proposed Feature Block Texture Difference (FBTD) which refers to the texture variation of facial patch, the final expressional feature are formed and inputted to One-dimensional Convolution Neural Network (1DCNN) for FER. Five experiments are conducted in this research, and three average FER rates 95.2%, 96.5%, and 97% for Beihang University (BHU) facial expression database, MMI facial expression database, and the combination of two databases, respectively, have shown the significant advantages of the proposed method over the existing ones.     

A hybrid multiple feature construction approach for classification using Genetic Programming

The purpose of feature construction is to create new higher-level features from original ones. Genetic Programming (GP) was usually employed to perform feature construction tasks due to its flexible representation. Filter-based approach and wrapper-based approach are two commonly used feature construction approaches according to their different evaluation functions. In this paper, we propose a hybrid feature construction approach using genetic programming (Hybrid-GPFC) that combines filter’s fitness function and wrapper’s fitness function, and propose a multiple feature construction method that stores top excellent individuals during a single GP run. Experiments on ten datasets show that our proposed multiple feature construction method (Fcm) can achieve better (or equivalent) classification performance than the single feature construction method (Fcs), and our Hybrid-GPFC can obtain better classification performance than filter-based feature construction approaches (Filter-GPFC) and wrapper-based feature construction approaches (Wrapper-GPFC) in most cases. Further investigations on combinations of constructed features and original features show that constructed features augmented with original features do not improve the classification performance comparing with constructed features only. The comparisons with three state-of-art methods show that in majority of cases, our proposed hybrid multiple feature construction approach can achieve better classification performance.     

PMU based voltage security assessment of power systems exploiting principal component analysis and decision trees

The need for feature selection and dimension reduction is felt as a fundamental step in security assessment of large power systems in which the number of features representing the state of power grids dramatically increases. These large amounts of attributes are not proper to be used for computational intelligence (CI) techniques as inputs, because it may lead to a time consuming procedure with insufficient results and they are not suitable for on-line purposes and updates.This paper proposes a combined method for an online voltage security assessment in which the dimension of the token data from phasor measurement units (PMUs) is reduced by principal component analysis (PCA). Then, the features with different stability indices are put into several categories and feature selection is done by correlation analysis in each category. These selected features are then given to decision trees (DTs) for classification and security assessment of power systems.The method is applied to 39-bus test system and a part of Iran power grid. It is seen from the results that the DTs with reduced data have simpler splitting rules, better performance in saving time, reasonable DT error and they are more suitable for constant updates.     

Stereo matching cost computation based on nonsubsampled contourlet transform

A new matching cost computation method based on nonsubsampled contourlet transform (NSCT) for stereo image matching is proposed in this paper. Firstly, stereo image is decomposed into high frequency sub-band images at different scales and along different directions by NSCT. Secondly, by utilizing coefficients in high frequency domain and grayscales in RGB color space, the computation model of weighted matching cost between two pixels is designed based on the gestalt laws. Lastly, two types of experiments are carried out with standard stereopairs in the Middlebury benchmark. One of the experiments is to confirm optimum values of NSCT scale and direction parameters, and the other is to compare proposed matching cost with nine known matching costs. Experimental results show that the optimum values of scale and direction parameters are respectively 2 and 3, and the matching accuracy of the proposed matching cost is twice higher than that of traditional NCC cost.     

基于子波域空间相关的多分辨图像滤波方法

从静态子波变换入手,提出了一种有效的图像滤波算法。通过计算相邻尺度下细节信号的空间相关性来区分噪声和信号,如果子波系数的空间相关性大,则认为此位置的系数含有特征及边缘信息予以保留。实现这种理论的完整方法包括两部分:空间滤波和子波收缩。仿真结果表明这个算法有很好的降噪性能。     

An intelligent web-page classifier with fair feature-subset selection

The explosion of on-line information has given rise to many manually constructed topic hierarchies (such as Yahoo!!). But with the current growth rate in the amount of information, manual classification in topic hierarchies results in an immense information bottleneck. Therefore, developing an automatic classifier is an urgent need. However, classifiers suffer from enormous dimensionality, since the dimensionality is determined by the number of distinct keywords in a document corpus. More seriously, most classifiers are either working slowly or they are constructed subjectively without any learning ability. In this paper, we address these problems with a fair feature-subset selection (FFSS) algorithm and an adaptive fuzzy learning network (AFLN) for classification. The FFSS algorithm is used to reduce the enormous dimensionality. It not only gives fair treatment to each category but also has ability to identify useful features, including both positive and negative features. On the other hand, the AFLN provides extremely fast learning ability to model the uncertain behavior for classification so as to correct the fuzzy matrix automatically. Experimental results show that both FFSS algorithm and the AFLN lead to a significant improvement in document classification, compared to alternative approaches.     

提高前馈神经网络推广能力的若干实际方法

提高前馈神经网络的推广能力是深受关注的问题，本文根据我们最近提出了一个网络有效推广的准则，从提高网络特征提取能力，分类能力和修改神经元激活函数等几方面给出了若干实际方案，我们在任意手写数字识别问题上的实验结果证实了这些方法的有效性。     

Merging Functional Form Feature into GT Codes to Enhance the CAD/CAM Integration

The TangShan Research Institute of Group Technology has been engaged in the develop-ment of GT for the light industry machinery building factories since 1980.Under its overall supervisionmore than 8 factories have applied GT to drawing control and process planning,while 4 of them reorga-nized their machine shops into GT cells and the throughput capabilities were raisd immediately by 15 toeven 49%.Since the existing GT coding systems predict only the global characteristics of machine parts,theirapplication alone is unable to transfer all the information needed for the CAD/CAM integration in aCIM environment.From the very beginning of 1989,we started to study the feasibility of introducingconcepts of Functional Form Features into GT codes.Based on the exhaustive study of nearly 100,000part drawings from machineries for making pottery and porcelain appliances,cigarette,paper pulp,glass bottles,plastics moulding,beer and beverage filling etc.,a new GT/FFF coding system wasworked out and a corresponding set of standard drawings assembled from FFF primitives Was formulat-ed which can cover up to 65—75% of existing parts from light industry machineries.FFF based GTcodes greatly facilitate the standardization control of released drawings,accelerate the engineering de-sign and process planning,and certainly will play an important role in the future CIM organization.