分布式光纤入侵信号检测与识别

分布式光纤声波传感(DAS)技术通过接收相干瑞利散射光的相位信息来探测声波或振动信号,具有灵敏度高、动态范围广等特性,可利用线性定量测量实现对信号的高保真还原.随着实际应用的需求不断提高,光纤入侵检测领域对事件的定位和识别提出了更高的要求,表现为对入侵事件的准确分类,因此将分布式光纤声波传感技术与模式识别(PR)技术相...     

被动声纳目标识别系统设计研究

目标识别是声纳的主要功能之一,其性能包括识别正确率、泛化能力和识别距离。由于数据样本的保密特性,声纳目标识别系统设计有其自身特色。在设计过程中,应首先降低对数据样本的依赖,把目标辐射噪声的机理分析、组成、运动规律等知识巧妙地运用到系统设计中;其次还应尽可能提高泛化率,以提高声纳目标识别系统对未见过的样本的正确识别能力。讨论了声纳目标识别流程、声纳目标识别系统的设计方法;在声纳目标识别系统的性能评估中,给出了目标识别距离的估算方法。     

A comprehensive study on face recognition: methods and challenges

ABSTRACT

Face Recognition is the process of identifying and verifying the faces. Face recognition has vast importance in the field of Security, Healthcare, Banking, Criminal Identification, Payment, and Advertising. In this paper, we have reviewed various techniques and challenges for the face recognition. Illumination, pose variation, facial expressions, occlusions, aging, etc. are the key challenges to the success of face recognition. Pre-processing, Face Detection, Feature Extraction, Optimal Feature Selection, and Classification are primary steps in any face recognition system. This paper provides a detailed review of each. Feature extraction techniques can be classified as appearance-based methods or geometry-based methods, such method may be local or global. Feature extraction is the most crucial stage for the success of the face recognition system. However, deep learning methods have freed the user from handcrafting the features. In this article, we have surveyed state-of-the-art methods of last few decades and the comparative study of various feature extraction methods is provided. Article also describes the current challenges in the area.     

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。     

基于特征选择和GWO-KELM的鸟声识别算法

针对鸟声识别算法中提取特征单一、分类准确率低等问题,提出一种基于混合特征选择和灰狼算法优化核极限学习机的鸟声识别方法。首先从鸟声数据中提取大规模声学特征集ComParE,其次计算每个特征的F_score并进行排序,然后以广义顺序向前浮动搜索(Generalized Sequential Forward Floating Search, GSFFS)为搜索策略,特征子集在核极限学习机(Kernel Limit Learning Machine, KELM)上十折交叉验证的正确率,作为特征选择标准进行特征选择,得到适用于鸟声识别的特征子集,最后通过灰狼算法(Grey Wolf Optimizer, GWO)选择最优KELM参数识别鸟声。在柏林自然科学博物馆鸟声数据库中进行实验,该方法在60类鸟声识别平均正确率和F1-score达到94.45%和92.29%。结果表明,该方法相较于传统自行设计提取的单一特征集具有更高的识别精度,GWO-KELM模型比网格搜索方式更易找到全局最优值。     

Food recognition using neural network classifier and multiple hypotheses image segmentation

ABSTRACT

This paper proposes the multiple-hypotheses image segmentation and feed-forward neural network classifier for food recognition to improve the performance. Initially, the food or meal image is given as input. Then, the segmentation is applied to identify the regions, where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the features of every food item are extracted by the global feature and local feature extraction. After the features are obtained, the classification is performed for each segmented region using a feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food volume and (ii) calorie and nutrition measure based on mass value. The experimental results and performance evaluation are validated. The outcome of the proposed method attains 0.947 for Macro Average Accuracy (MAA) and 0.959 for Standard Accuracy (SA), which provides better classification performance.     

Comparative study of feature extraction and classification methods for recognition of characters taken from vehicle registration plates

ABSTRACT

General Optical Character Recognition system works on the base of several successive steps such as pre-processing, segmentation, feature extraction, classification and post-processing. Feature extraction plays here a major role. In this article, we present an overview and comparison of various methods and approaches for off-line recognition of machine written Latin characters. We assume that individual characters are already segmented in an image. To recognize characters and translate them to text requires that each character must be described by a feature vector, which is then classified into one of the 36 classes corresponding to the uppercase Latin alphabet letters and numbers.     

Pattern recognition of jet fuels: comprehensive GC×GC with ANOVA-based feature selection and principal component analysis

Two-dimensional comprehensive gas chromatography (GC×GC) is applied to a pattern recognition problem involving classification of jet fuel mixtures. Analysis of variance (ANOVA)-based feature selection is initially used to identify and select chromatographic features relevant to a given classification in two studies. Then, principal component analysis (PCA) was used for pattern recognition classification. In the first study, a 1% volumetric composition change in mixtures of JP-5 and JP-7 jet fuel is readily distinguished. In this first study, the effective combination of GC×GC, ANOVA-based feature selection and PCA is developed and evaluated as a chemical analysis tool. The second study involved the analysis of three samples each of three different jet fuel types, JP-5, JP-8, and JP-TS, as well as blends incorporating two or three jet fuels. Each of the nine jet fuel samples originated from various geographic locations within the United States. These samples were analyzed in order to determine if a classification based on fuel type is possible in the presence of sample variability (due to geographic origin) with GC×GC/pattern recognition analysis. Chromatographic features that are adept at classification of jet fuel type and are not sensitive to geographic origin of the sample were generated for the sample set consisting of the original fuel types as well as mixtures of the three different, original jet fuels. The combination of GC×GC with ANOVA-based feature selection was found to be a useful tool to enhance the chemical selectivity, and thus the classification power of the analytical procedure, when coupled with PCA.     

基于模拟加载系统油压信号的自动测试与识别技术研究

在液压系统模拟加载与自动测试、识别过程中,工作装置油压波动信号是一种典型的非平稳信号。针对其影响因素多、不具备明显频域特征以及任何单一特征参量都无法对信号进行准确识别的难题,提出了对信号先进行状态分割,在分割基础上计算不同工作状态下的特征参量,并进行基于主成分分析(PCA)的特征提取方法,最后采用最小二乘支持向量机(LSSVM)构建多分类器,实现对工作装置6种不同工作状态的准确识别。实验结果验证了该方法的有效性,为同类液压系统的信号特征分析及模式识别提供了参考。