Gabor-based kernel PCA for palmprint recognition

A novel Gabor-based kernel principal component analysis (PCA) method by integrating the Gabor wavelet representation of palm images and the kernel PCA method for palmprint recognition is proposed. The feasibility of the proposed method has been successfully tested on two different public data sets from the PolyU palmprint databases, for which the samples were collected in two different sessions.     

Verification system robust to occlusion using low-order Zernike moments of palmprint sub-images

This paper proposes a palmprint based verification system which uses low-order Zernike moments of palmprint sub-images. Euclidean distance is used to match the Zernike moments of corresponding sub-images of query and enrolled palmprints. These matching scores of sub-images are fused using a weighted fusion strategy. The proposed system can also classify the sub-image of palmprint into non-occluded or occluded region and verify user with the help of non-occluded regions. So it is robust to occlusion. The palmprint is extracted from the acquired hand image using a low cost flat bed scanner. A palmprint extraction procedure which is robust to hand translation and rotation on the scanner has been proposed. The system is tested on IITK, PolyU and CASIA databases of size 549, 5239 and 7752 hand images respectively. It performs with accuracy of more than 98%, and FAR, FRR less than 2% for all the databases.     

Personal recognition using hand shape and texture.

This paper proposes a new bimodal biometric system using feature-level fusion of hand shape and palm texture. The proposed combination is of significance since both the palmprint and hand-shape images are proposed to be extracted from the single hand image acquired from a digital camera. Several new hand-shape features that can be used to represent the hand shape and improve the performance are investigated. The new approach for palmprint recognition using discrete cosine transform coefficients, which can be directly obtained from the camera hardware, is demonstrated. None of the prior work on hand-shape or palmprint recognition has given any attention on the critical issue of feature selection. Our experimental results demonstrate that while majority of palmprint or hand-shape features are useful in predicting the subjects identity, only a small subset of these features are necessary in practice for building an accurate model for identification. The comparison and combination of proposed features is evaluated on the diverse classification schemes; naive Bayes (normal, estimated, multinomial), decision trees (C4.5, LMT), k-NN, SVM, and FFN. Although more work remains to be done, our results to date indicate that the combination of selected hand-shape and palmprint features constitutes a promising addition to the biometrics-based personal recognition systems.     

基于灰度分布特征的层次掌纹识别方法

目前掌纹识别算法主要集中在对掌纹图像所切取的ROI区域的研究,而对原始手部图像的灰度分布特征则讨论较少。在一定光照条件下,掌部不同位置的主线、皱纹和表层皮肤颜色的深浅在灰度图像上各自对应了不同的灰度级。该文提出一种利用手部尺寸和角度信息完成粗分类,借助单元信息熵的概念来分析手部图像的灰度分布特征从而完成细分类的层次掌纹识别方法。该方法不同于传统的对ROI区域进行特征提取的方法,直接利用整幅人手图像完成分类识别。在99类共990幅手部图像的数据库上进行的实验结果与PCA和LDA算法的对比表明,该算法具有比传统算法更高的鲁棒性,识别率也得到了较大幅度的提高。     

基于曲面类型与深度学习融合的三维掌纹识别技术

传统的2维掌纹识别在图像采集时容易受到干湿度、残影和压力等影响,使得其鲁棒性和准确性降低。为解决这些问题,3维掌纹识别技术应运而生。现有的3维掌纹身份认证技术需要将掌纹的特征提取与匹配识别分开进行,不仅延缓了识别时间,更增加了不同方法优化组合的难度。该文提出一种基于曲面类型(ST)与深度学习融合的3维掌纹识别方法。该方法利用ST图像表示3维掌纹特征,并将其作为卷积神经网络(CNN)的输入,实现网络的训练。测试图像可自行提取掌纹图像特征信息并在网络中直接完成识别。实验结果表明,该文方法在公开数据集上得到了99.43%的准确率和28 ms的识别时间,与传统3维掌纹识别方法相比均有提高,实现了3维掌纹的快速高精度识别。     

MEEMD特征掌纹的2DPCA识别方法

为了提高识别率,提出了基于MEEMD和2DPCA的掌纹识别方法.利用MEEMD技术对掌纹图像进行分解,得到本征模式函数（IMF）分量,用高频分量重构掌纹图像,形成掌纹识别图像集.然后利用2DPCA技术进行识别.MEEMD重构掌纹能够突出掌纹细节特征,提高识别率.采用香港理工大学掌纹数据库进行实验,将此方法与不包含MEEMD的2DPCA方法进行比较,实验结果说明此方法有较高的识别率和较快的识别速度.     

人体手掌静脉图像采集系统研究

以手掌静脉图像采集装置为研究对象,详细分析了采集装置的设计原理,设计了一种光强分布均匀的双环形近红外光源结构,并提出了一种基于灰度共生矩阵及模糊逻辑的手掌静脉图像质量评价方法。通过研究对比度、熵、相关性的值与静脉纹理结构信息的关系,建立模糊推理系统对图像质量做出了最佳判别,从而验证了采集系统的性能。设计的采集装置结构小、成本低,采集到的静脉图像比较清晰,满足静脉识别系统后续处理的需要。     

基于改进S变换的手掌静脉身份识别

针对目前基于手掌静脉的身份识别问题,分析了 掌脉图像的纹理特性,提出一种利用改进的S变换 能量特征的识别方法。首先对掌脉图像进行二维离散正交S变换(2D-DOST),将变换后的 频域图像作为 掌脉纹理特征的一种体现,即形成S变换能量曲面(STES);然后将不同掌脉能量曲面进行 减法运算,得 到能量差曲面,进一步计算曲面的标准差,并以此为依据对不同掌脉进行分类识别。在通 用的接触式标 准掌脉图库上测试的结果表明,本文方法正确识别率为99.892,识 别时间为0.017s;同时使用自主 设计的采集仪在非接触环境下测试的结果表明,本文方法正确识别率可达99.480,识别时间为 0.020s。相比其他典型和流行算法,本文算 法提高了手掌静脉识别系统的性能,具有可行性和有效性。     

基于DSP的掌纹图像定位算法的实现

文中详细阐述一套基于DSP的掌纹定位算法的实现过程。采用形态学的方法提取掌纹图像的关键点,运用＂错切＂原理算法对图像进行旋转,具有很高的准确性和运行效率。文中的方法简单有效,但也具有一定的局限性。     

一种新的掌纹ROI图像定位方法

在分析传统Harris算法缺陷的基础上,提出一种基于角点检测的快速掌纹图像预处理方法。该方法利用一种改进的自适应Harris算法提取出手掌边缘和手指凹凸处的轮廓特征点,采用属于同一特征组的特征点均值作为候选角点,剔除了邻近角点,并根据候选角点在轮廓线上角度变化的大小定位手指根部的三个关键角点,从而有效提取出掌纹图像的感兴趣区域(ROI)。实验测试图像采用CASIA数据库,实验结果表明,该方法能快速有效地定位出掌纹的ROI区域,且能有效增强掌纹图像的边缘与细节,有利于提高掌纹识别的识别率。     

基于核主元分析和Fisher线性判别的掌纹识别

提出了基于核主元分析(KPCA)和FLD相结合的掌纹识别方法.对每幅掌纹图像应用KPCA进行降维,然后将二维图像矩阵转换成一维图像矢量.PolyU掌纹图像库中所有图像矢量组成的数据矩阵作为FLD的输入,进行特征提取,计算特征矢量间的余弦距离进行掌纹匹配.实验结果说明,与传统的PCA+FLD相比,在不同的特征个数下,本文方法均取得了较小的等错率(EER),而且特征提取时间较短,运行速度较快.在三种不同的核函数中,RBF核函数的识别效果最佳,等错率最小为0.     

基于离散曲波变换和支持向量机的掌纹识别方法

提出了一种基于离散曲波变换和支持向量机的掌纹识别方法.首先将所有掌纹样本图像和测试图像通过基于Wrapping的快速离散曲波变换进行分解,从而获得不同尺度、不同角度的曲波变换系数;掌纹重要特征信息包含在曲波变换分解系数中的低频系数中,因此将分解系数变换形成特征向量后作为特征参数送入支持向量机中进行学习训练;最后将训练好的支持向量机用于掌纹分类.基于香港理工大学Palmprint掌纹数据库进行了大量实验,实验结果证实所提方法的识别正确率相对优于小波变换方法和其它几种经典方法.     

基于波长选择的手掌静脉采集系统研究

手掌静脉成像是手掌静脉识别的难点,设计了一种在拍摄全手的情况下能够采集到较清晰手掌静脉图像的低成本多波段采集系统,并分别在760 nm,850 nm,890 nm和960 nm四种单波长及两两组合的近红外光源下采集手掌静脉,总样本数为2500幅手掌静脉图像,然后利用提取的有效区域中手掌静脉平均灰度与非手掌静脉平均灰度的比值为波长选择标准对图像质量进行比较,最后选择在760 nm和960 nm混合波长下采集手掌静脉。     

Palmprint and Finger-Knuckle-Print for efficient person recognition based on Log-Gabor filter response

Person recognition systems based on biometrics are being increasingly utilized in any applications to enhance the security of physical and logical access systems. A number of biometric traits exist and are in use in various applications. Each biometric trait has its strengths and weaknesses, and the choice depends on the application. Palmprint is one of the relatively new biometrics due to its stable and unique characteristics. The rich texture information of palmprint offers one of the powerful means in person recognition. An important issue in palmprint recognition is to extract features that can discriminate an individual from the other. Recently, a novel hand-based biometric feature, Finger-Knuckle-Print (FKP), has attracted an increasing amount of attention. Like any other biometric identifiers, FKPs are believed to have the critical properties of universality, uniqueness and permanence for person recognition. In this paper, we propose a multiple traits system for person recognition using palmprint and FKP. We have used 1D Log-Gabor response to extract the information from these two traits. So, each trait is represented by the real and the imaginary templates. Such extracted templates are compared with those of the database using the Hamming distance. Using the Hong Kong Polytechnic University (PolyU) database, the experimental results showed that the proposed system achieves excellent performances in terms of computation cost and of recognition rates, for both verification and identification.     

基于双Gabor方向韦伯局部描述子的掌纹识别

该文结合掌纹图像的纹理特点,对原始韦伯局部描述子(WLD)中的差分激励和梯度方向进行改进,提出双Gabor方向韦伯局部描述子(DGWLD),以提高掌纹识别率。在构建新的差分激励图时,通过加入邻域像素点与中心像素点间灰度差分的方向信息,扩大异类掌纹间的差异。同时,采用双Gabor方向代替原始的梯度方向,减小平移和旋转对识别的影响。此外,为了更好地衡量特征间的相似度,使用交叉匹配算法,进一步提升识别率。在PolyU, MSpalmprint和CASIA掌纹库上进行实验,识别率均达到100%。实验的结果表明,与其它局部描述子和已有改进的WLD方法相比,该文方法具有更高的识别率和更低的等错误率。     

一种掌纹纹线结构特征的描述和匹配方法

掌纹的纹线结构特征明显,稳定且具有足够的辨别信息。由于掌纹线很不规则,很难用数学表达式的方式对其结构进行精确描述。该文提出一种新的用于掌纹识别的纹线结构特征描述和匹配方法。该方法应用改进的注水算法从二值的纹线图像上提取纹线的统计结构特征;标记纹线方向,提取纹线的方向结构特征;将掌纹线的全局统计结构特征和局部细节方向结构特征整合为一个特征向量描述掌纹纹线的结构特征;最后定义一个加权距离衡量不同纹线结构特征之间的相似度。实验表明,该方法不仅具有较好的匹配性能,而且可以获得较高的识别精度。     

An Adaptive Approach to Singular Point Detection in Fingerprint Images

This paper presents a novel algorithm for detection of singular points, the core and delta points, in fingerprint images. The number and location of singular points, are used to classify fingerprint images into five general groups; and therefore to narrow down the search space in large fingerprint databases. Using the proposed directional masks in the first step, we detect the neighborhood of the singular points. In the second stage by implementing the proposed algorithm, an adaptive singular point detection method, is designed to extract the exact location of core and delta points. Usage of the proposed directional masks speeds up the process and the proposed adaptive singular point detection method increases the accuracy of the algorithm.     

基于核局部Fisher判别分析的掌纹识别

运用核局部Fisher判别分析(KLFDA)进行掌纹识别。为了解决小样本图像识别中特征方程矩阵的奇异性问题,首先运用图像下抽样方法降低掌纹空间的维数,在低维图像上应用KLF-DA提取低维的投影向量;然后将训练图像和待识别图像的核矩阵向投影向量上投影,得到非线性局部判别特征;最后计算特征向量间的余弦距离,进行掌纹匹配。运用PolyU掌纹图像库对算法进行测试,实验结果表明,与主元分析(PCA)、Fisher判别分析(FDA)、独立元分析(ICA)、核主元分析(KPCA)和局部Fisher判别分析(LFDA)相比,本文算法的识别率(RR)最高为99%,特征提取和匹配总时间0.031 s,满足实时系统的要求。