基于步态特征的身份识别算法研究

步态识别根据人走路的姿势进行身份识别,由于人在行走时在空间呈现出的不同几何模式,单一几何特征难以全面描述步态特征,导致身份识别正确率不高。为提高身份识别的正确率,提出一种空间和频率特征模式相融合的身份识别算法。首先利用摄像机采集步态图像序列,然后分别采用极坐系和傅里叶变换提取步态空间特征和频率特征,并对两种特征进行融合,最后采用支持向量机对融合特征进行学习和分类,进行身份识别。结果表明,相对于单一步态特征为参数的身份识别算法,融合算法的身份识别正确率有了明显提高,且具有更好的稳定性。     

基于模糊支持向量机的步态身份识别研究

由于传统支持向量机本身一些固有的缺陷,众多的学者开始将模糊数学的思想引入支持向量机中,在传统支持向量机的基础上加入了“模糊隶属度”因子．从而构造出了一种新的分类器一模糊支持向量机。本文力图通过分析模糊支持向量机在语言识别方面已有的实验成果。探讨模糊支持向量机在步态识别中的可行性,从而期望模糊支持向量机在步态识别领域能够取得更好的分类效果。     

基于心电脉搏特征的视觉疲劳状态识别

为从生物医学信号角度检测和评估视觉疲劳,模拟VDT作业环境,对35位健康被试者进行1.5 h的VDT疲劳实验。使用MP425数据采集卡和LabVIEW构成的数据采集系统同步采集心电(ECG)和脉搏波信号,经信号预处理分析后,提取实验前后的ECG和脉搏波信号特征。研究结果表明,ECG和脉搏波信号特征在实验前后有较大变化,采用支持向量机法对实验前后的ECG脉搏组合特征进行分类,正确率可达100%。     

基于自适应稀疏分解的声音识别算法

针对公共环境中异常声音的检测与识别存在的强噪声干扰及检测效率低的问题,提出基于参数自适应匹配跟踪的声信号识别算法.基于粒子和种群的进化率改进粒子群参数的自适应设置并优化稀疏分解目标函数;基于自适应粒子群算法的连续集搜索特性建立连续超完备Gabor原子集,以提高最匹配优原子与声信号的匹配度并加速原子的匹配搜索;使用SVM分类器实现公共环境异常声信号的复合特征识别.实验结果表明,与已有算法相比,该算法的公共环境异常声信号的识别率最优,且对不同背景噪声具有较好的识别鲁棒性.     

应用于动态心电监测的自动身份识别系统

心电信号由于其易于监测和重要的临床应用价值成为物联网获取健康信息的重要内容。人体心电信号虽然是非稳态时变信号,但现有研究表明其可以作为一种新型的生物识别特征用于对识别分辨率要求不高的应用场合。本项目开发了一套应用于动态心电监测的自动用户身份识别系统。该系统侧重在手机端利用接收到的心电信号实现自动用户身份识别,并在远程接入服务器时基于身份识别结果将心电数据存入对应的用户数据库。所开发的身份识别系统可加载于现有加密模块之上,从而实现保密加认证的双重安全需求。     

基于支持向量机算法的气体识别研究

利用多传感器或者传感器阵列,同时,结合神经网络技术来进行气体识别和定量分析研究已成为目前传感器领域的一个研究热点。介绍了一种在该领域还没有引起足够重视的算法———支持向量机算法(SVM)。利用该算法,结合多传感器技术,对 3种不同体积分数的有机溶剂进行了识别研究,并取得了较好的识别效果,证明了该算法在气体识别领域具有相当大的研究价值。     

基于动态心电信号的实时身份识别算法

心电图(ECG)信号因其具备易于监测、个体唯一性等特点在生物识别领域受到广泛关注.针对身份识别的准确性和实时性问题,给出一种快速鲁棒的、适用于微型化嵌入式平台的心电信号身份识别算法.首先,利用动态阈值法提取稳定波形用于快速生成心电模板样本和测试样本;然后,基于优化动态时间弯曲(DTW)法进行差异度计算得到识别结果;其次,考虑心电信号为非稳态时变信号,为保证模板数据与人体体征状况的一致性,对心电模板库进行动态更新管理以进一步提高识别准确性与鲁棒性.对MIT-BIH心律失常数据库和自建心电数据库的分析结果表明:所述算法的识别成功率最高达到98.6%;在安卓移动端,动态阈值与优化DTW法一次运算平均时间分别约为59.5 ms和26.0 ms,实时性能显著提高.     

基于连续波多普勒雷达的手势信号识别算法

手势识别技术在人机交互系统中的需求与应用日益广泛,毫米波雷达可以对手势运动过程中的距离、速度信息进行检测,从而实现识别的目的,且具有不依赖于光线的优点。利用24 GHz微型雷达装置可接收手势信号,建立识别算法的样本库。对一维手势信号进行分段FFT运算,可将一维手势信号转化为二维的手势图像,转换后的图像不但含有运动过程中的幅度、速度信息,还包括手势运动过程中幅度与频谱的变化历程。由于每一次手势动作的不确定性,单一的物理特征统计方法很难进行判别,且识别率较差,利用机器学习SVM算法对手势信号进行学习与分类。实验结果表明,分段FFT信号处理方法结合SVM算法对手势分类的准确率达90. 25%,为手势识别算法提供了一种新的方法。     

基于支持向量机算法的葡萄酒种类识别

本文研究了一种基于支持向量机的葡萄酒分类模型,选取UCI数据库中的wine数据集进行检验.结果显示,将支持向量机算法引入葡萄酒分类模型,可以减少参数选取的盲目性,提高预测分类的准确性.     

基于改进多核学习的语音情感识别算法

提出一种基于改进多核学习的语音情感识别算法.算法以高斯径向基核函数为基准,通过采样不同的样本,采用不同的评价标准并获得不同的参数,来提高分类性能.此外,通过引入多核技术,将得到的高斯核函数构建多核学习的基核,并通过利用松弛因子构建的软间隔多核学习的目标函数改善了学习效率.对比仿真实验结果表明,本文提出的基于多核学习语音情感识别算法有效提高了语音情感识别性能.     

一种改进的稀疏度自适应匹配追踪算法

在信号稀疏度未知的情况下,稀疏度自适应匹配追踪算法(Sparsity Adaptive Matching Pursuit,SAMP)是一种广泛应用的压缩感知重构算法。为了优化SAMP算法的性能,提出了一种改进的稀疏度自适应匹配追踪(Improved Sparsity Adaptive Matching Pursuit,ISAMP)算法。该算法引入广义Dice系数匹配准则,能更准确地从测量矩阵中挑选与残差信号最匹配的原子,利用阈值方法选取预选集,并在迭代过程中采用指数变步长。实验结果表明,在相同的条件下,改进后的算法提高了重构质量和运算速度。     

基于压缩感知信号重建的自适应空间正交匹配追踪算法

传统的奈奎斯特采样定理规定采样频率最少是原信号频率的两倍,才能保证不失真的重构原始信号,而压缩感知理论指出只要信号具有稀疏性或可压缩性,就可以通过采集少量信号来精确重建原始信号.在研究和总结已有匹配算法的基础上,提出了一种新的自适应空间正交匹配追踪算法(Adaptive Space Orthogonal Matching Pursuit,ASOMP)用于稀疏信号的重建.该算法在选择原子匹配时采用逆向思路,引入正则化自适应和空间匹配的原则,加快了原子的匹配速度,提高了匹配的准确性,最终实现了原始信号的精确重建.最后与传统MP和OMP算法进行了仿真对比,结果表明该算法的重建质量和算法速度均优于传统MP和OMP算法.     

Metacarpophalangeal joint patterns based personal identification system

This paper introduces a novel approach for identity authentication system based on metacarpophalangeal joint patterns (MJPs). A discriminative common vector (DCV) based method is utilized for feature selection. In the literature, there is no study using whole MJP for identity authentication, exceptionally a work (Ferrer et al., 2005) using the hand knuckle pattern which is some part of the MJP draws the attention as a similar study. The originality of this approach is that: whole MJP is firstly used as a biometric identifier and DCV method is firstly applied for extracting the feature set of MJP. The developed system performs some basic tasks like image acquisition, image pre-processing, feature extraction, matching, and performance evaluation. The feasibility and effectiveness of this approach is rigorously evaluated using the k-fold cross validation technique on two different databases: a publicly available database and a specially established database. The experimental results indicate that the MJPs are very distinctive biometric identifiers and can be securely used in biometric identification and verification systems, DCV method is successfully employed for obtaining the feature set of MJPs and proposed MJP based authentication approach is very successful according to state of the art techniques with a recognition rate of between 95.33% and 100.00%.     

用于CS的广义稀疏度自适应匹配追踪算法

压缩感知理论的基本思想是原始信号在某一变换域是稀疏的或者是可压缩的,并将奈奎斯特采样定理中的采样过程和压缩过程合二为一。稀疏度自适应匹配追踪（SAMP）算法能够实现稀疏度未知情况下的重构,而广义正交匹配追踪算法每次迭代时选择多个原子,提高了算法的收敛速度。基于上述两种重构算法的优势,提出了广义稀疏度自适应匹配追踪（Generalized Sparse Adaptive Matching Pursuit,gSAMP）算法。针对重构图像的峰值信噪比、重构时间、相对误差等客观评价指标,以及主观视觉上对所提算法与传统的贪婪算法进行对比。在压缩比固定为0.5时,gSAMP算法的重构效果优于传统的MP、OMP、ROMP、SAMP以及gOMP贪婪类重构算法的效果。     

Electroencephalogram subject identification: A review

This is, to the best of the authors knowledge, the first complete research on the state of the art on EEG based subject identification. As well as covering the full story of this field (from 1980 to 2013), an overview of the findings made in genetic and neurophysiology areas, from which it is based, is also provided. After a comprehensive search, 109 biometric publications were found and studied, from which 88 were finally included in this document. A categorization of papers is proposed based on the recording paradigm. The most used databases, some of them public, have been identified and named to allow the comparison of results from these and future works. The findings of this work show that, although basic questions remain to be answered, the EEG, and specially its power spectrum in the range of the alpha rhythm, contains subject specific information that can be used for classification. Moreover, approaches such as a multi-day-session training, the fusion of information from different electrodes and bands, and Support Vector Machines are recommended to maximize the system’s performance. All in all, the problem of subject identification by means of their EEG is harder than initially expected, as it relies on information extracted from complex heterogeneous EEG traits which are the results of elaborated models of inheritance, which in turn makes the problem very sensitive to its variables (time, frequency, space, recording paradigm and algorithms).     

ECG to identify individuals

The electrocardiogram (ECG also called EKG) trace expresses cardiac features that are unique to an individual. The ECG processing followed a logical series of experiments with quantifiable metrics. Data filters were designed based upon the observed noise sources. Fiducial points were identified on the filtered data and extracted digitally for each heartbeat. From the fiducial points, stable features were computed that characterize the uniqueness of an individual. The tests show that the extracted features are independent of sensor location, invariant to the individual's state of anxiety, and unique to an individual.     

基于回溯思想的多面体面追踪算法

多面体面追踪算法能有效求解基追踪算法(BP)的对偶问题,但是算法一步只能选择一个原子,算法效率比较低.为解决上述问题,采用回溯迭代的思想对多面体面追踪算法进行改进,改进后的稀疏度自适应的多面体面追踪算法一步可以选择多个原子,同时利用回溯思想将可信度较低的原子删除,不但提高了算法的速度和重构的精度,而且实现了对信号稀疏度的自适应.通过仿真证明改进后的多面体面追踪算法的重构效率明显优于多面体面追踪算法,而且重构时间明显降低.     

MISO-FIR系统的梯度追踪辨识算法

针对含有未知时滞的多输入单输出有限脉冲响应系统,根据系统参数化后具有的稀疏特性,基于压缩感知原理,将匹配追踪方法和梯度搜索原理相结合,在有限采样数据下,提出了可以同时估计系统参数和时滞的梯度追踪算法．该算法同正交匹配追踪算法相比,梯度追踪算法具有较小的计算量．最后通过仿真验证了算法的有效性．     

分块子空间追踪算法

压缩传感理论是一种充分利用信号稀疏性或者可压缩性的全新信号采样理论。该理论表明,通过采集少量的信号测量值就能够实现可稀疏信号的精确重构。本文在研究现有经典重构算法的基础上,提出结合图像分块思想和回溯思想的分块子空间追踪算法（Block Subspace Pursuit, B_SP）用于压缩传感信号的重构。该算法以块结构获取图像,利用回溯过程实现支撑集的自适应筛选,最终实现图像信号的精确重构。实验结果表明,在相同测试条件下,该算法的重构效果无论从主观视觉上还是客观数据上都有不同程度的提高。     

Personal identification based on finger-vein features

Finger-vein recognition refers to a recent biometric technique which exploits the vein patterns in the human finger to identify individuals. The advantages of finger vein over traditional biometrics (e.g. face, fingerprint, and iris) lie in low-risk forgery, noninvasiveness, and noncontact. This paper here presents a new method of personal identification based on finger-vein recognition. First, a stable region representing finger-vein network is cropped from the image plane of an imaging sensor. A bank of Gabor filters is then used to exploit the finger-vein characteristics at different orientations and scales. Based on the filtered image, both local and global finger-vein features are extracted to construct a finger-vein code (FVCode). Finally, finger-vein recognition is implemented using the cosine similarity measure classifier, and a fusion scheme in decision level is adopted to improve the reliability of identification. Experimental results show that the proposed method exhibit an exciting performance in personal identification.