Application of wavelet transform in feature extraction and pattern recognition of wideband echoes

A novel approach to extract edge features from wideband echo is proposed. The set of extracted features not only represents the echo waveform in a concise way, but also is sufficient and well suited for classification of non-stationary echo data from objects with different property.The feature extraction is derived from the Discrete Dyadic Wavlet Transform (DDWT) of the echo through the undecimated algorithm. The motivation we use the DDWT is that it is time-shift-invariant which is beneficial for localization of edge, and the wavelet coefficients at larger scale represent the main shape feature of echo, i.e. edge, and the noise and modulated high-frequency components are reduced with scale increased. Some experimental results using real data which contain 144 samples from 4 classes of lake bottoms with different sediments are provided. The results show that our approach is a prospective way to represent wideband echo for reliable recognition of nonstationary echo with great variability.

Pattern recognitionbased method for radar antideceptive jamming

In order to make the effective ECCM to the deceptive jamming, especially the angle deceptive jamming, this paper establishes a signal-processing model for anti-deceptive jamming firstly, in which two feature-extracting algorithms, i.e. the statistical algorithm and the neural network （NN） algorithm are presented, then uses the RBF NN as the classitier in the processing model. Finally the two algorithms are validated and compared through some simulations.

Feature extraction of partial discharge in low-temperature composite insulation based on VMD-MSE-IF

Low-temperature composite insulation is commonly applied in high-temperature super-conducting apparatus while partial discharge(PD)is found to be an important indicator to reveal insulation statues.In order to extract feature parameters of PD signals more effectively,a method combined variational mode decomposition with multi-scale entropy and image feature is proposed.Based on the simulated test platform,original and noisy signals of three typical PD defects were obtained and decomposed.Accordingly,relative moments and grayscale co-occurrence matrix were employed for feature extraction by K-modal component diagram.Afterwards,new PD feature vectors were obtained by dimension reduction.Finally,effectiveness of different feature extraction methods was evaluated by pattern recognition based on support vector machine and K-nearest neighbour.Result shows that the proposed feature extraction method has a higher recognition rate by comparison and is robust in processing noisy signals.

New supervised learning classifiers for structural damage diagnosis using time series features from a new feature extraction technique

The motivation for this article is to propose new damage classifiers based on a supervised learning problem for locating and quantifying damage.A new feature extraction approach using time series analysis is introduced to extract damage-sensitive features from auto-regressive models.This approach sets out to improve current feature extraction techniques in the context of time series modeling.The coefficients and residuals of the AR model obtained from the proposed approach are selected as the main features and are applied to the proposed supervised learning classifiers that are categorized as coefficient-based and residual-based classifiers.These classifiers compute the relative errors in the extracted features between the undamaged and damaged states.Eventually,the abilities of the proposed methods to localize and quantify single and multiple damage scenarios are verified by applying experimental data for a laboratory frame and a four-story steel structure.Comparative analyses are performed to validate the superiority of the proposed methods over some existing techniques.Results show that the proposed classifiers,with the aid of extracted features from the proposed feature extraction approach,are able to locate and quantify damage;however,the residual-based classifiers yield better results than the coefficient-based classifiers.Moreover,these methods are superior to some classical techniques.

Pattern Recognition of Modulation Signal Classification Using Deep Neural Networks

In recent times,pattern recognition of communication modulation signals has gained significant attention in several application areas such as military,civilian field,etc.It becomes essential to design a safe and robust feature extraction(FE)approach to efficiently identify the various signal modulation types in a complex platform.Several works have derived new techniques to extract the feature parameters namely instant features,fractal features,and so on.In addition,machine learning(ML)and deep learning(DL)approaches can be commonly employed for modulation signal classification.In this view,this paper designs pattern recognition of communication signal modulation using fractal features with deep neural networks(CSM-FFDNN).The goal of the CSM-FFDNN model is to classify the different types of digitally modulated signals.The proposed CSM-FFDNN model involves two major processes namely FE and classification.The proposed model uses Sevcik Fractal Dimension(SFD)technique to extract the fractal features from the digital modulated signals.Besides,the extracted features are fed into the DNN model for modulation signal classification.To improve the classification performance of the DNN model,a barnacles mating optimizer(BMO)is used for the hyperparameter tuning of the DNN model in such a way that the DNN performance can be raised.A wide range of simulations takes place to highlight the enhanced performance of the CSM-FFDNN model.The experimental outcomes pointed out the superior recognition rate of the CSM-FFDNN model over the recent state of art methods interms of different evaluation parameters.

Robust Feature Extraction for Speaker Recognition Based on Constrained Nonnegative Tensor Factorization

How to extract robust feature is an important research topic in machine learning community. In this paper, we investigate robust feature extraction for speech signal based on tensor structure and develop a new method called constrained Nonnegative Tensor Factorization （cNTF）. A novel feature extraction framework based on the cortical representation in primary auditory cortex （A1） is proposed for robust speaker recognition. Motivated by the neural firing rates model in A1, the speech signal first is represented as a general higher order tensor, cNTF is used to learn the basis functions from multiple interrelated feature subspaces and find a robust sparse representation for speech signal. Computer simulations are given to evaluate the performance of our method and comparisons with existing speaker recognition methods are also provided. The experimental results demonstrate that the proposed method achieves higher recognition accuracy in noisy environment.

Finger crease pattern recognition using Legendre moments and principal component analysis

The finger joint lines defined as finger creases and its distribution can identify a person. In this paper, we propose a new finger crease pattern recognition method based on Legendre moments and principal component analysis （PCA）. After obtaining the region of interest （ROI） for each finger image in the pre- processing stage, Legendre moments under Radon transform are applied to construct a moment feature matrix from the ROI, which greatly decreases the dimensionality of ROI and can represent principal components of the finger creases quite well. Then, an approach to finger crease pattern recognition is designed based on Karhunen-Loeve （K-L） transform. The method applies PCA to a moment feature matrix rather than the original image matrix to achieve the feature vector. The proposed method has been tested on a database of 824 images from 103 individuals using the nearest neighbor classifier. The accuracy up to 98.584% has been obtained when using 4 samples per class for training. The experimental results demonstrate that our proposed approach is feasible and effective in biometrics.

Research on Intelligent Identification of PD Patterns Based on the Fingerprint Features

Five-electrode configurations were designed to simulate the distribution inhomogeneity of electric field intensities in the air-insulating medium, and the characteristic data waveforms of partial discharge generated by different electrode configurations under the excitation of power frequency AC voltage were carefully collected in this paper. Furthermore, the feature vectors of the corresponding fingerprint, contained in partial discharge data, were extracted by rigorous mathematical algorithms, and the artificial neural network was employed to realize the pattern recognition of partial discharge caused by the inhomogeneity of electric field intensity with different electrode configurations. The results indicate that the J4 value in the space of 7 feature quantities is 1905.6, and the recognition rate is 100% when the hidden layer neuron of the network is 19. However, the J5 value of 9 feature quantities is 1589.9, and the purpose of recognition has been achieved when the number of hidden layer neurons of the network is 6. Increasing the number of hidden layer neurons will only waste computing resources. Of course, PD information collection mode, feature quantity selection, optimal feature space composition, network structure and classification algorithm are the key to realizing PD fault intelligence identification.

An Improved Real-Time Face Recognition System at Low Resolution Based on Local Binary Pattern Histogram Algorithm and CLAHE

This research presents an improved real-time face recognition system at a low resolution of 15 pixels with pose and emotion and resolution variations. We have designed our datasets named LRD200 and LRD100, which have been used for training and classification. The face detection part uses the Viola-Jones algorithm, and the face recognition part receives the face image from the face detection part to process it using the Local Binary Pattern Histogram (LBPH) algorithm with preprocessing using contrast limited adaptive histogram equalization (CLAHE) and face alignment. The face database in this system can be updated via our custom-built standalone android app and automatic restarting of the training and recognition process with an updated database. Using our proposed algorithm, a real-time face recognition accuracy of 78.40% at 15 px and 98.05% at 45 px have been achieved using the LRD200 database containing 200 images per person. With 100 images per person in the database (LRD100) the achieved accuracies are 60.60% at 15 px and 95% at 45 px respectively. A facial deflection of about 30° on either side from the front face showed an average face recognition precision of 72.25%-81.85%. This face recognition system can be employed for law enforcement purposes, where the surveillance camera captures a low-resolution image because of the distance of a person from the camera. It can also be used as a surveillance system in airports, bus stations, etc., to reduce the risk of possible criminal threats.

RESEARCH ON TWO-DIMENSIONAL LDA FOR FACE RECOGNITION

The letter presents an improved two-dimensional linear discriminant analysis method for feature extraction. Compared with the current two-dimensional methods for feature extraction, the improved two-dimensional linear discriminant analysis method makes full use of not only the row and the column direc-tion information of face images but also the discriminant information among different classes. The method is evaluated using the Nanjing University of Science and Technology (NUST) 603 face database and the Aleix Martinez and Robert Benavente (AR) face database. Experimental results show that the method in the letter is feasible and effective.

RESEARCH ON FACE RECOGNITION BASED ON IMED AND 2DPCA

This letter proposes an effective method for recognizing face images by combining two-Dimen- sional Principal Component Analysis (2DPCA) with IMage Euclidean Distance (IMED) method. The proposed method is comprised of four main stages. The first stage uses the wavelet decomposition to extract low fre- quency subimages from original face images and omits the other three subimages. The second stage concerns the application of IMED to face images. In the third stage, 2DPCA is employed to extract the face features from the processed results in the second stage. Finally, Support Vector Machine (SVM) is applied to classify the extracted face features. Experimental results on the AR face image database show that the proposed method yields better recognition performance in comparison with the 2DPCA method that is not combined with IMED.

Fabric Pattern Elements Retrieval Based on Cosine Transform

Fabric pattern contains many types of the available pattern elements, which not only can be used for the researchers, but also as the material for the designer. But existing method focus on the complete image retrieval, therefore lack methods of retrieving pattern elements. This article proposes a pattern elements retrieval algorithm based on cosine transform. Firstly, automatically segment the patterns according to size and location and filter the similar primary patterns, then, through cosine transform, analyze elements features in DCT domain, extract amplitude frequency and phase frequency. We employ 2-norm to measure the similarity, search 10 similar pattern elements in the sample library and save them in the design resources library. Experiment results indicate that this algorithm performs well while used in palace costume and carpet patterns, and got more than 75% of the average recall in 100 times experiments

Study on the Essence of Optimal Statistically Uncorrelated Discriminant Vectors and Its Application to Face Recognition

A study has been made on the essence of optimal uncorrelated discriminant vectors. A whitening transform has been constructed by means of the eigen decomposition of the population scatter matrix, which makes the population scatter matrix be an identity matrix in the transformed sample space no matter whether the population scatter matrix is singular or not. Thus, the optimal discriminant vectors solved by the conventional linear discriminant analysis (LDA) methods are statistically uncorrelated. The research indicates that the essence of the statistically uncorrelated discriminant transform is the whitening transform plus conventional linear discriminant transform. The distinguished characteristics of the proposed method is that the obtained optimal discriminant vectors are not only orthogonal but also statistically uncorrelated. The proposed method is applicable to all the problems of algebraic feature extraction. The numerical experiments on several facial databases show the effectiveness of the proposed method.

基于FrFT和RVM的变压器局部放电模式识别

快速准确识别局部放电类型对于保证变压器安全稳定运行具有重要意义。针对局部放电信号模式识别中面临的最优特征参数提取和分类器设计难题,提出一种基于分数阶傅里叶变换(fractional Fourier transform,FrFT)和相关向量机(relevance vector machine,RVM)的局部放电模式识别方法。首先将FrFT引入局部放电信号分析领域,利用FrFT将局部放电信号转换至分数域并对其进行多尺度分析,在扩充信息提取维度的同时,提取可反映不同局部放电信号波形差异的14维特征构成特征向量;然后将特征向量作为输入,建立RVM模型进行最优特征选择和分类判决函数的联合优化,从而实现对不同局部放电信号的分类识别。建立电晕放电、沿面放电和气隙放电试验模型并采集局部放电超声信号开展试验,结果表明所提方法对于每种局部放电信号均能获得较高的识别精度,平均正确识别率相对于常规支持向量机(support vector machine,SVM)分类方法提升超过2.7%。

Application of Computer Vision Technique to Maize Variety Identification

Variety identification is important for maize breeding, processing and trade. The computer vision technique has been widely applied to maize variety identification. In this paper, computer vision technique has been summarized from the following technical aspects including image acquisition, image processing, characteristic parameter extraction, pattern recognition and programming softwares. In addition, the existing problems during the application of this technique to maize variety identification have also been analyzed and its development tendency is forecasted.

参数化服装样板生成技术研究进展

基于参数化设计的服装样板生成技术是服装制版领域的研究热点。为促进参数化设计在服装个性化制版中的研究与发展,系统介绍国内外人体特征识别与参数提取、服装参数化样板规则构建及制版系统设计等方面的研究进展;重点分析在2D、3D空间进行人体特征识别及参数提取的研究,其中3D空间相较于2D空间,保留了人体的内在信息,有利于从多个角度有效识别人体特征,且效率较高。结合服装数字化、智能化的发展趋势,指出在研究样板生成规则的同时,提高特征参数提取的效率和准确性,进行参数化智能制版系统设计是未来研究的主要方向。

麦克风集群网络模型及其障碍物识别中的应用

基于声学基本原理和抽象空间的网络结构方法,开发了一种创新的障碍物检测系统,其结构包括硬件终端、多声路网络模型和分类器。硬件终端由立体式分布的麦克风节点、数据采集卡和工控机组成。多声路网络模型能够基于多路声波信号建立网络结构,并提取和计算多路声波复杂信号中的有效信息。分类器采用主成分分析方法(Principal Component Analysis,PCA)对提取的有效信息进行降维处理,再采用监督学习方法实现对环境的分类和识别。该系统的开发不仅为汽车自动驾驶、机器人导航和无人机等领域提供了重要的技术支持,而且为智能导航和自动化领域的技术发展提供了新的可能性。未来的工作将进一步优化系统性能,提高其应用的广泛性和经济效益。

MFCC特征训练技术在声纹识别中的应用

阐述MFCC声纹特征提取的原理、MFCC特征提取模式和基于MFCC声纹识别的实现。提取模式包括MFCC提取流程、短时傅立叶变换STFT、梅尔滤波器组的构造、离散余弦变换(DCT)与MFCC特征值的提取。针对融合特征提取方案可分性与鲁棒性的缺陷提出改进方案。

基于CEEMD-MPE与SDAE的局部放电模式识别

针对变压器局部放电故障信息提取困难以及局部放电类型识别准确率低等问题,提出一种基于CEEMD-MPE与SDAE相结合的局部放电模式识别算法。对局部放电原始信号进行CEEMD分解,得到多个固有模态分量(IMF),根据相关系数筛选出系数最大的IMF作为最优分量,计算其不同尺度下的排列熵值;将有效排列熵值作为特征数据集输入到SDAE中进行无监督学习训练;利用Softmax分类器输出放电类型。实验结果表明,该算法识别精准率为98%,召回率为96.67%,F1得分为97.17%,能够快速、准确地识别局部放电类型。

基于GA-BP的表面肌电信号下肢动作模式识别研究

为了满足下肢康复机器人运动过程中对人体下肢不同动作模式的识别的需求,首先,通过8通道无线肌电传感器采集8种下肢常见动作的表面肌电(sEMG)信号,并对原始信号进行滤波、运动段提取、特征提取处理;然后,将处理后数据分别输入本文设计的BP、PCA-BP、GA-BP、PCA-GA-BP分类器进行训练与测试。4种分类器对下肢8种动作平均识别率分别为88.6%,90.5%,92.3%,95.1%,对每个动作平均识别率为85%以上。结果表明:基于GA-BP神经网络比BP神经网络具有更高的预测精度,并且降维处理可以提高动作分类的准确率。