Detection and classification of buried dielectric anomalies bymeans of the bispectrum method and neural networks

The development of a neural network-based system for detection and classification of buried landmines is the main focus of this paper. Shape-dependent features are extracted by means of the bispectrum method. These features are then applied to the neural network. A multilayer back-propagation-type neural network is trained and tested on the feature sets extracted from equally spaced radial slices of image windows. Simulation results obtained for two types of targets indicated good detection and classification rates     

Bridging semantic gap between high-level and low-level features in content-based video retrieval using multi-stage ESN–SVM classifier

Content-based video retrieval system aims at assisting a user to retrieve targeted video sequence in a large database. Most of the search engines use textual annotations to retrieve videos. These types of engines offer a low-level abstraction while the user seeks high-level semantics. Bridging this type of semantic gap in video retrieval remains an important challenge. In this paper, colour, texture and shapes are considered to be low-level features and motion is a high-level feature. Colour histograms convert the RGB colour space into YcbCr and extract hue and saturation values from frames. After colour extraction, filter mask is applied and gradient value is computed. Gradient and threshold values are compared to draw the edge map. Edges are smoothed for sharpening to remove the unnecessary connected components. These diverse shapes are then extracted and stored in shape feature vectors. Finally, an SVM classifier is used for classification of low-level features. For high-level features, depth images are extracted for motion feature identification and classification is done via echo state neural networks (ESN). ESN are a supervised learning technique and follow the principle of recurrent neural networks. ESN are well known for time series classification and also proved their effective performance in gesture detection. By combining the existing algorithms, a high-performance multimedia event detection system is constructed. The effectiveness and efficiency of proposed event detection mechanism is validated using MSR 3D action pair dataset. Experimental results show that the detection accuracy of proposed combination is better than those of other algorithms     

An Optimized Approach to Vehicle-Type Classification Using a Convolutional Neural Network

Vehicle type classification is considered a central part of an intelligent traffic system. In recent years, deep learning had a vital role in object detection in many computer vision tasks. To learn high-level deep features and semantics, deep learning offers powerful tools to address problems in traditional architectures of handcrafted feature-extraction techniques. Unlike other algorithms using handcrated visual features, convolutional neural network is able to automatically learn good features of vehicle type classification. This study develops an optimized automatic surveillance and auditing system to detect and classify vehicles of different categories. Transfer learning is used to quickly learn the features by recording a small number of training images from vehicle frontal view images. The proposed system employs extensive data-augmentation techniques for effective training while avoiding the problem of data shortage. In order to capture rich and discriminative information of vehicles, the convolutional neural network is fine-tuned for the classification of vehicle types using the augmented data. The network extracts the feature maps from the entire dataset and generates a label for each object (vehicle) in an image, which can help in vehicle-type detection and classification. Experimental results on a public dataset and our own dataset demonstrated that the proposed method is quite effective in detection and classification of different types of vehicles. The experimental results show that the proposed model achieves 96.04% accuracy on vehicle type classification.     

Classification of audio radar signals using radial basis function neural networks

Radial basis function (RBF) neural networks are used to classify real-life audio radar signals that are collected by a ground surveillance radar mounted on a tank. Currently, a human operator is required to operate the radar system to discern among signals bouncing off tanks, vehicles, planes, and so on. The objective of this project is to investigate the possibility of using a neural network to perform this target recognition task, with the aim of reducing the number of personnel required in a tank. Different signal classification methods in the neural net literature are considered. The first method employs a linear autoregressive (AR) model to extract linear features of the audio data, and then perform classification on these features, i.e, the AR coefficients. AR coefficient estimations based on least squares and higher order statistics are considered in this study. The second approach uses nonlinear predictors to model the audio data and then classifies the signals according to the prediction errors. The real-life audio radar data set used here was collected by an AN/PPS-15 ground surveillance radar and consists of 13 different target classes, which include men marching, a man walking, airplanes, a man crawling, and boats, etc. It is found that each classification method has some classes which are difficult to classify. Overall, the AR feature extraction approach is most effective and has a correct classification rate of 88% for the training data and 67% for data not used for training.     

A VIBRATION RECOGNITION METHOD BASED ON DEEP LEARNING AND SIGNAL PROCESSING

Effective vibration recognition can improve the performance of vibration control and structural damage detection and is in high demand for signal processing and advanced classification.Signal-processing methods can extract the potent time-frequency-domain characteristics of signals;however,the performance of conventional characteristics-based classification needs to be improved.Widely used deep learning algorithms(e.g.,convolutional neural networks(CNNs))can conduct classification by extracting high-dimensional data features,with outstanding performance.Hence,combining the advantages of signal processing and deep-learning algorithms can significantly enhance vibration recognition performance.A novel vibration recognition method based on signal processing and deep neural networks is proposed herein.First,environmental vibration signals are collected;then,signal processing is conducted to obtain the coefficient matrices of the time-frequency-domain characteristics using three typical algorithms:the wavelet transform,Hilbert-Huang transform,and Mel frequency cepstral coefficient extraction method.Subsequently,CNNs,long short-term memory(LSTM)networks,and combined deep CNN-LSTM networks are trained for vibration recognition,according to the time-frequencydomain characteristics.Finally,the performance of the trained deep neural networks is evaluated and validated.The results confirm the effectiveness of the proposed vibration recognition method combining signal preprocessing and deep learning.     

基于深度神经网络的货架商品识别方法

目的为快速统计货架商品信息,提出一种基于深度神经网络的货架商品自动识别方法。方法摄像头采集的货架商品图像经过深度神经网络算法处理,得到了图像中商品的SKU和位置。针对货架商品识别这种密集检测场景,文中方法改进了通用深度神经网络目标检测算法,将算法分为检测和分类2个阶段且重新设计了部分网络结构。最后,将文中方法和传统货架商品识别方法以及通用深度神经网络目标检测方法进行了比较。结果实验证明该方法的检测阶段的模型平均正确率达到96.5%,分类阶段的分类准确率达到99.9%。整图测试的查准率为97.56%,查全率为99.26%。结论相较于以往使用传统的目标检测模型进行货架商品识别以及使用SIFT等人工算子提取特征并分类识别商品具体SKU,文中方法的商品检出率和分类准确率都有了大幅度的提升,具有很好的应用潜力。     

基于卷积神经网络模型的遥感图像分类

研究了遥感图像的分类,针对遥感图像的支持向量机(SVM)等浅层结构分类模型特征提取困难、分类精度不理想等问题,设计了一种卷积神经网络(CNN)模型,该模型包含输入层、卷积层、全连接层以及输出层,采用Soft Max分类器进行分类。选取2010年6月6日Landsat TM5富锦市遥感图像为数据源进行了分类实验,实验表明该模型采用多层卷积池化层能够有效地提取非线性、不变的地物特征,有利于图像分类和目标检测。针对所选取的影像,该模型分类精度达到94.57%,比支持向量机分类精度提高了5%,在遥感图像分类中具有更大的优势。     

冲击回波信号的波形结构特征提取及分类研究

针对典型的非均质材料混凝土内部缺陷大小无损检测这一难题,利用人的听觉系统的生理学和心理学的研究成果,基于ＥＩＨ模型建立了进行时域波形结构特征提取的方法,并用结构自适应模糊聚类神经网络对提取的特征进行了分类试验,结果表明该特征提取方法能较好地解决从冲击反射时域信号来进行肃离缺陷大小识别问题。     

Analysis and classification of malignancy in pancreatic magnetic resonance images using neural network techniques

Computer-aided diagnosis (CAD) is a computerized way of detecting tumors in MR images. Magnetic resonance imaging (MRI) has been generally used in the diagnosis and detection of pancreatic tumors. In a medical imaging system, soft tissue contrast and noninvasiveness are clear preferences of MRI. Inaccurate detection of tumor and long time consumption are the disadvantages of MRI. Computerized classifiers can greatly renew the diagnosis activity, in terms of both accuracy and time necessity by normal and abnormal images, automatically. This article presents an intelligent, automatic, accurate, and robust method to classify human pancreas MRI images as normal or abnormal in terms of pancreatic tumor. It represents the response of artificial neural network (ANN) and support vector machine (SVM) techniques for pancreatic tumor classification. For this, we extract features from MR images of pancreas using the GLCM method and select the best features using JAFER algorithm. These features are analyzed by five classification techniques: ANN BP, ANN RBF, SVM Linear, SVM Poly, and SVM RBF. We compare the results with benchmark data set of MR brain images. The analytical outcome presents that the two best features used to classify the MR images using ANN BP technique have 98% classification accuracy.     

Accurate Multi-Scale Feature Fusion CNN for Time Series Classification in Smart Factory

Time series classification (TSC) has attracted various attention in the community of machine learning and data mining and has many successful applications such as fault detection and product identification in the process of building a smart factory. However, it is still challenging for the efficiency and accuracy of classification due to complexity, multi-dimension of time series. This paper presents a new approach for time series classification based on convolutional neural networks (CNN). The proposed method contains three parts: short-time gap feature extraction, multi-scale local feature learning, and global feature learning. In the process of short-time gap feature extraction, large kernel filters are employed to extract the features within the short-time gap from the raw time series. Then, a multi-scale feature extraction technique is applied in the process of multi-scale local feature learning to obtain detailed representations. The global convolution operation with giant stride is to obtain a robust and global feature representation. The comprehension features used for classifying are a fusion of short time gap feature representations, local multi-scale feature representations, and global feature representations. To test the efficiency of the proposed method named multi-scale feature fusion convolutional neural networks (MSFFCNN), we designed, trained MSFFCNN on some public sensors, device, and simulated control time series data sets. The comparative studies indicate our proposed MSFFCNN outperforms other alternatives, and we also provided a detailed analysis of the proposed MSFFCNN.     

High impedance fault detection in power distribution networks using time?frequency transform and probabilistic neural network

An intelligent approach for high impedance fault (HIF) detection in power distribution feeders using advanced signal-processing techniques such as time-time and time-frequency transforms combined with neural network is presented. As the detection of HIFs is generally difficult by the conventional over-current relays, both time and frequency information are required to be extracted to detect and classify HIF from no fault (NF). In the proposed approach, S- and TT-transforms are used to extract time-frequency and time-time distributions of the HIF and NF signals, respectively. The features extracted using S- and TT-transforms are used to train and test the probabilistic neural network (PNN) for an accurate classification of HIF from NF. A qualitative comparison is made between the HIF classification results obtained from feed forward neural network and PNN with same features as inputs. As the combined signal-processing techniques and PNN take one cycle for HIF identification from the fault inception, the proposed approach was found to be the most suitable for HIF classification in power distribution networks with wide variations in operating conditions.     

An automated methodology for the classification of focal and nonfocal EEG signals using a hybrid classification approach

The uncertainty in human brain leads to the formation of epilepsy disease in human. The automatic detection and severity analysis of epilepsy disease is proposed in this article using a hybrid classification algorithm. The proposed method consists of decomposition stage, feature extraction, and classification stages. The electroencephalogram (EEG) signals are decomposed using dual-tree complex wavelet transform and then features are extracted from these coefficients. These features are then classified using the neural network classification approach in order to classify the EEG signals into either focal or nonfocal EEG signals. Furthermore, severity of the focal EEG signal is analyzed using an adaptive neuro-fuzzy inference system classification approach. The proposed hybrid classification method for the classification of focal signals and nonfocal signals achieved 98.6% of sensitivity, 99.1% of specificity, and 99.4% of accuracy. The average detection rate for both focal and nonfocal dataset is about 98.5%.     

基于融合特征的短语音汉语声调自动识别方法

提出一种基于韵律特征(基频、时长)和梅尔倒谱系数(Mel-Frequency Cepstral Coefficient,MFCC)特征的融合特征进行短语音汉语声调识别的方法,旨在利用两种特征的优势提高短语音汉语声调识别率。该融合特征包括7个根据不同模型得到的韵律特征和统计参数以及4个从每个音段的梅尔倒谱系数计算得来的对数化后验概率,使用高斯混合模型表示4个声调的倒谱特征的分布。实验分两步:第一步,将基于韵律特征和倒谱特征的分类器在决策阶段混合起来进行声调分类,分别赋予两个分类器权重,计算倒谱特征和韵律特征在声调分类任务中的权重;第二步,将基于字的韵律特征和基于帧的倒谱特征结合起来生成融合特征的超向量,使用融合特征进行汉语声调识别,根据准确率、未加权平均召回率(Unweigted Average Recall,UAR)和科恩卡帕(Cohen’s Kappa)系数3个指标,比较并评估5种分类器(两种设置的高斯混合模型,后向传播神经网络,支持向量机和卷积神经网络(Convolutional Neural Network,CNN))在不平衡数据集上的分类效果。实验结果表明:(1)倒谱特征方法能够提高汉语声调的识别率,该特征在总体分类任务中的权重为0.11;(2)基于融合特征的深度学习(CNN)方法对声调的识别率最高,为87.6%,与高斯混合模型的基线系统相比,提高了5.87%。该研究证明了倒谱特征法能够提供与韵律特征法互补的信息,从而提高短语音汉语声调识别率;同时,该方法可以运用到韵律检测和副语言信息检测等相关研究中。     

基于深度度量学习的轴承故障诊断方法

针对机械大数据因故障类内离散度和类间相似度较大而导致诊断精度低的问题,提出一种深度度量学习故障诊断方法,采用深度神经网络(Deep Neural Network, DNN)对故障特征进行自适应提取,并利用基于欧氏距离的边际Fisher分析(Marginal Fisher Analysis, MFA)方法进行了优选,在构建的深度度量网络(Deep Metric Network, DMN)顶层特征输出层添加BPNN(Back Propagation Neural Network, BPNN)分类器对网络参数进行微调,并实现故障的分类识别。通过对不同类型和严重程度的轴承故障进行了诊断分析,验证了该方法可以有效地对轴承故障进行高精度诊断,效果优于传统深度信念网络(Deep Belief Network, DBN)故障诊断方法以及常用时域统计特征结合支持向量机(Support Vector Machine, SVM)分类的故障诊断方法。     

基于多分辨率时频特征融合的声学场景分类

声学场景分类是计算机听觉中最难的任务之一,在单一特征条件下采用基本的卷积神经网络相对于传统的分类方法精度已经有所提升,但是效果依然不够理想。针对这一问题,在卷积神经网络框架下,提出了一种基于时频特征融合的声学场景分类方案。在分类模型构建方面,提出一种多分辨率卷积池化方案,构造多分辨率卷积神经网络,以更好地适应提取特征的时频结构;在特征选取方面,融合低层次包络特征对数——Mel子带能量和高层次结构特征——非负矩阵分解系数矩阵,把两种二维特征堆叠为三维特征送入分类模型。在2017年和2018年声学场景分类和事件检测挑战赛的开发数据集上进行了训练和测试。实验结果表明,文中提出方案比基线系统的分类精度分别提高7.5%和10.3%,可有效改善分类效果。     

ECG Classification Using Deep CNN Improved by Wavelet Transform

Atrial fibrillation is the most common persistent form of arrhythmia. A method based on wavelet transform combined with deep convolutional neural network is applied for automatic classification of electrocardiograms. Since the ECG signal is easily inferred, the ECG signal is decomposed into 9 kinds of subsignals with different frequency scales by wavelet function, and then wavelet reconstruction is carried out after segmented filtering to eliminate the influence of noise. A 24-layer convolution neural network is used to extract the hierarchical features by convolution kernels of different sizes, and finally the softmax classifier is used to classify them. This paper applies this method of the ECG data set provided by the 2017 PhysioNet/CINC challenge. After cross validation, this method can obtain 87.1% accuracy and the F1 score is 86.46%. Compared with the existing classification method, our proposed algorithm has higher accuracy and generalization ability for ECG signal data classification.     

A fuzzy logic‐based meningioma tumor detection in magnetic resonance brain images using CANFIS and U‐Net CNN classification

This article develops a methodology for meningioma brain tumor detection process using fuzzy logic based enhancement and co‐active adaptive neuro fuzzy inference system and U‐Net convolutional neural network classification methods. The proposed meningioma tumor detection process consists of the following stages as, enhancement, feature extraction, and classifications. The enhancement of the source brain image is done using fuzzy logic and then dual tree‐complex wavelet transform is applied to this enhanced image at different levels of scale. The features are computed from the decomposed sub band images and these features are further classified using CANFIS classification method which identifies the meningioma brain image from nonmeningioma brain image. The performance of the proposed meningioma brain tumor detection and segmentation system is analyzed in terms of sensitivity, specificity, segmentation accuracy, and Dice coefficient index with detection rate.     

基于小波包分解和PCA-Attention-LSTM的舰船辐射噪声识别技术

为了改进舰船辐射噪声分类系统的性能,进一步提高识别准确率,文章提出了一种基于多特征的小波包分解在长短期记忆(LongShort-TermMemory,LSTM)网络中分类的方法。该方法首先通过小波包分解技术,分频段提取舰船辐射噪声的多种特征,将提取的特征利用主成分分析法(Principal Component Analysis, PCA)进行数据降维,通过添加注意力机制(Attention Mechanism)算法的LSTM网络,对辐射噪声结果分类,提高了学习效率和识别准确率。为了更精细地提取特征,分频段提取了舰船辐射噪声的时频域特征、小波变换特征和梅尔倒谱系数等特征,并将分频段与不分频段的特征、多特征与单一特征、不同信噪比间的算法性能进行对比。实验结果表明,基于小波包分解和PCA-Attention-LSTM的模型可以有效地提高舰船辐射噪声分类的性能,是一种可行的分类方法。     

基于特征融合进行活动识别的DCNN 5法

研究了输入是可穿戴传感器获得的多通道时间序列信号,输出是预定义的活动的活动识别模型,指出活动中的有效特征的提取目前多依赖于手工和浅层特征学习结构,不仅复杂而且会导致识别准确率下降;基于深度学习的卷积神经网络（ CNN）不是对时间序列信号进行手工特征提取,而是自动学习最优特征;目前使用卷积神经网络处理有限标签数据仍存在过拟合问题。因此提出了一种基于融合特征的系统性的特征学习方法用于活动识别,用ImageNet16对原始数据集进行预训练,将得到的数据与原始数据进行融合,并将融合数据和对应的标签送入有监督的深度卷积神经网络（ DCNN ）中,训练新的系统。在该系统中,特征学习和分类是相互加强的,它不仅能处理端到端的有限数据问题,也能使学习到的特征有更强的辨别力。与其他方法相比,该方法整体精度从87.0％提高到87.4％。     

应用卷积神经网络的锂电池极片涂布缺陷分类

目的 针对锂电池极片涂布缺陷种类多,传统方法分类检测精度不高,以及人工依赖性强等问题,提出一种基于卷积神经网络的锂电池极片涂布缺陷自动分类算法。方法 首先对网络结构以及模型参数进行优化,接着在网络中加入跳跃连接结构,将空洞卷积提取到的多尺度特征与高层特征进行融合以获取更多缺陷特征,并采用LeakyReLU(Leaky Rectified Linear Unit)激活函数保留图像中的负值特征信息,最后通过构建的数据集训练模型,实现锂电池极片涂布缺陷的准确分类。结果 实验结果表明,当前方法识别准确率能够达到99.34%,平均检测时间为51ms。结论 改进后的方法能够准确分类出锂电池极片18种涂布缺陷,满足工业生产中实时分类检测的要求。