共查询到17条相似文献,搜索用时 93 毫秒
1.
2.
3.
基于独立成分分析的掌纹识别 总被引:6,自引:0,他引:6
本文研究了独立成分分析(ICA)两种不同的结构ICA I和ICAII在掌纹识别中的应用.为了提高识别准确性和可靠性,该方法首先对掌纹图像进行预处理,提取掌纹感兴趣(ROI)区域进行特征提取和匹配.为了减少计算量,运用ICA算法之前,先采用主成分分析(PCA)算法去除掌纹图像的二阶统计特征相关性,其余的高阶统计特征由ICA分离.对于PolyU掌纹图像库,基于ICA模型的预测误差平方和(SPE)小于PCA,而且重构的原始图像优于PCA.为了比较两种算法识别性能,本丈分别用PCA、ICA I、ICAII提取特征掌纹子空间,然后将待识别图像投影到低维子空间上,最后用余弦距离进行掌纹匹配.实验结果表明,ICA算法两种结构的识别率均高于PCA,ICAII在性能上优于ICA I. 相似文献
4.
本文研究了独立成分分析(ICA)两种不同的结构ICAⅠ和ICAⅡ在掌纹识别中的应用。为了提高识别准确性和可靠性,该方法首先对掌纹图像进行预处理,提取掌纹感兴趣(ROI)区域进行特征提取和匹配。为了减少计算量,运用ICA算法之前.先采用主成分分析(PCA)算法去除掌纹图像的二阶统计特征相关性,其余的高阶统计特征由ICA分离。对于PolyU掌纹图像库,基于ICA模型的预测误差平方和(SPE)小于PCA,而且重构的原始图像优于PCA。为了比较两种算法识别性能,本文分别用PCA、ICAⅠ、ICAⅡ提取特征掌纹子空间,然后将待识别图像投影到低维子空间上,最后用余弦距离进行掌纹匹配。实验结果表明,ICA算法两种结构的识别率均高于PCA,ICAⅡ在性能上优于ICAⅠ。 相似文献
5.
运用改进的遗传算法进行盲信号分离,解决了算法的局部收敛问题。通过改进遗传算法的交叉操作,极大地提高遗传算法的收敛速度;强迫分离矩阵的各个行向量正交,解决算法的奇异性问题。仿真试验表明,该算法有很好的分离效果。 相似文献
6.
分离矩阵的学习算法是盲信号分离的关键技术,矩阵联合对角化的预白化JADE算法是一种基于四阶累计量的学习算法。本文简要介绍了JADE算法的基本原理,通过实例,采用JADE算法对盲信号进行分离。实验表明,JADE算法在盲源信号分离中是一种很有潜力的方法。 相似文献
7.
8.
提出了一种新的虹膜特征提取与识别方法,该方法利用核主成分分析(KPCA)在高维空间具有较强的特征选择能力来提取虹膜图像的纹理特征。采用了一种距离度量和支持向量机相结合的两级分类方法,前级采用欧式距离来度量图像间的相似性,若符合条件,给出分类结果,否则拒绝,并转入后一级分类器——支持向量机分类,以减少进入支持向量机的样本数目,该组合分类方法充分利用了支持向量机识别率高和距离度量速度快的优点。实验结果表明,该方法提高了虹膜识别率,是一种有效的虹膜识别方法。 相似文献
9.
基于独立分量分析的形状识别 总被引:1,自引:0,他引:1
物体的形状识别是模式识别的重要方向之一,广泛应用于图像分析、机器视觉和目标识别等领域。在介绍利用信号的高阶统计信息的独立分量分析方法基础上,提出了基于独立分量分析的形状识别方法。利用独立分量分析算法提取出图像的独立基,根据待识别图像在独立基上投影系数的差别进行分类识别。仿真实验结果表明,该方法对于形状识别有较高的识别率。 相似文献
10.
11.
Shu-Kai S. Fan Yen Lin Chih-Min Fan Yi-Yi Wang 《Chemometrics and Intelligent Laboratory Systems》2009,99(1):19-29
A novel component analysis model is proposed to identify the mixed process signals which are frequently encountered in the statistical process control (SPC) and engineering process control (EPC) practice. Based upon one of existing state-of-the-art evolutionary algorithms, called particle swarm optimization (PSO), the proposed model provides a solution (i.e., demixing matrix) by maximizing the determinant of the corresponding second-order moment (variance–covariance) matrix of the reconstructed signals. Then, the estimated demixing matrix is used to separate mixed signals arising from several original process signals. The process signals considered in this paper include inconsistent variance series, autoregressive (AR) series, step change, and Gaussian noises in the process data. In practice, most of industrial manufacturing processes can be well characterized by a mixture of these four types of data. By following the proposed model, the blind signal separation framework can be cast into a nonlinear constrained optimization problem, where only the demixing matrix appears as unknown. Several illustrative examples involving linear mixtures of the process signals with different statistical characteristics are demonstrated to justify the new component analysis model. 相似文献
12.
13.
提出一种基于独立分量分析与相关系数的机械故障特征提取方法。首先对不同工况的机械振动信号分别进行独立分量分析,获得各种工况信号的独立分量,这些独立分量中蕴含了该工况的一些内在特征;接着利用样本与不同工况信号提取的独立分量的相关系数绝对值的和作为该样本的特征,与直接利用相关系数作为特征相比鲁棒性与区分程度都得到提高;最后使用支持向量机作为分类器进行识别。分别进行了齿轮故障特征提取与轴承故障特征提取实验,实验结果表明,此方法可以很好地提取机械故障特征信息。本文方法的优点在于直接从振动信号的原始数据中进行特征提取,获取机械故障蕴含的一些特征,应用范围广,具有较高地工程应用价值。 相似文献
14.
独立分量分析方法能够将线性混合信号进行分离,得到统计独立的源信号,能用于提取组合语音的特征基函数。倒谱矢量符合ICA变换的假设条件,用ICA方法对MFCC特征进行转换得到ICA特征基,继而用于说话人识别,建立了一个基于独立分量分析的说话人识别系统。实验结果表明,在噪声环境下此系统具有更高的识别率。 相似文献
15.
16.
Mingjin Zhang Peijin TongWenming Wang Jinpei GengYiping Du 《Chemometrics and Intelligent Laboratory Systems》2011,105(2):207-214
A strategy based on Independent Component Analysis (ICA) and Uncorrelated linear discriminant analysis (ULDA) was proposed for proteomic profile analysis and potential biomarker discovery from proteomic mass spectra of cancer and control samples. The method mainly includes 3 steps: (1) ICA decomposition for the mass spectra; (2) selection of discriminatory independent components (ICs) using nonparametric Mann-Whitney U-test; and (3) selection of special peaks (m/z locations) as potential biomarkers by executing of ULDA on a mass spectra data set which was reconstructed with the m/z locations that collected from the selected discriminatory ICs. A colorectal cancer data set and an ovarian cancer data set were analyzed with the proposed method. As results, 9 and 10 m/z locations were selected as potential biomarkers for the colorectal and ovarian cancer data set respectively. The classification results of ULDA using the selected potential biomarkers yielded better results than fisher discriminant analysis (FDA) and principal component analysis (PCA), and could distinguish the disease samples from healthy controls on the independent test sets with 100% of sensitivities and specificities for the colorectal cancer dataset and 100% of sensitivity and 96.77% of specificity for the ovarian cancer dataset. 相似文献