基于Gabor滤波器的指纹图像快速增强

研究并实现了利用Gabor滤波器对指纹图像增强的算法。改进了指纹图像方向图和纹线频率的提取方法,同时也对Gabor滤波器的快速实现方法进行研究,首先把Gabor滤波器分解为多个不同方向上有着不同参数的一维高斯滤波器的组合,然后通过递归的方法分别实现这些高斯滤波器,最后实现了Gabor滤波器的快速算法在指纹图像增强中的应用。实践表明,该方法效果良好,速度快,能大幅度提高指纹图像质量。     

一种改进的基于Gabor滤波的指纹增强算法

研究并实现了基于Gabor滤波器的指纹增强算法,对纹线方向提取算法进行了改进,获得了较准确的方向图。并改进了纹线频率的提取方法,有效地避免了Gabor滤波后在脊线中留下伪纹线和小孔的问题。同时也对滤波算法进行了优化,有效地减少了运算量。实验证明,该方法工作稳定,效果良好,鲁棒性强,对低质量图像具有显著的增强效果。     

基于Gabor滤波的指纹图像增强算法

由于指纹的唯一性和不变性,指纹识别已成为当前最流行、最方便、最可靠的个人身份认证技术之一。指纹识别一般包括指纹采集、图像预处理、特征提取及特征匹配等几个步骤。其中图像预处理中的图像增强是最为关键的环节,直接影响特征提取与特征匹配。提出了一种基于Gabor滤波的指纹图像增强算法。阐述了Gabor滤波器的定义及其在指纹图像增强中的应用。对指纹图像方向图提取方法和频率计算方法作了改进尝试。实践表明,该方法是有效的、实用的。     

基于改进的Gabor滤波指纹图像增强算法研究

对Gabor滤波器进行改进,改进的算法利用了指纹图像的局部特性,结合局部四邻城的关联特性,并且采用固定频率来代替复杂的频率计算.实验结果证明,改进的Gabor滤波算法不仅增强了指纹图像信息,同时也提高了处理速度.     

基于Gabor函数的小波域指纹图像增强算法

针对指纹大规模采集库中存在的指纹图像局部区过干或过湿的问题，提出了一种基于Gabor函数的小波域指纹增强算法。该算法在小波域利用低频系数图估计指纹方向，从而抑制了指纹局部过干或过湿的影响，进而分别实现基于Gabor函数的小波域各子图增强，最终将各增强子图利用小波逆变换实现重构。通过对FVC2004的DB1指纹库中的部分低质量图像的增强结果比较，该算法对低质量指纹图像的增强效果明显，且处理速度明显快于现存的Gabor增强方法。     

基于加博函数的指纹增强算法及其应用

指纹增强对于提高细节特征提取的准确率乃至整个自动指纹识别系统的性能实现都具有重要的意义.研究了基于加博(Gabor)函数的指纹增强算法,对纹线方向提取算法进行了改进,提出了一种纹线频率提取方法,给出了加博函数用于指纹增强的具体应用形式,以指纹图像的纹线方向和纹线频率为参数,使用加博函数实现了对指纹图像的增强处理,并使用南京大学活体指纹库中的部分典型低质量图像样本对算法性能进行了测试.实验结果表明,该算法对低质量图像具有显著的增强效果,增强后可以有效地降低指纹匹配的拒识率(false reject rate,简称FRR),处理速度和增强效果可以基本上满足在线自动指纹识别系统的需要.     

一种新型自适应指纹图像增强算法

针对Gabor滤波器在处理低质量指纹图像时增强效果不佳的缺陷,提出一种基于方向域和频率域的新型自适应Gabor滤波增强算法。该算法改进了指纹图像方向图和纹线频率的提取方法,通过计算图像方向一致性和平均纹线频率两个参数来共同自适应地调节Gobor滤波器变换窗的大小。实验结果表明：对于低质量指纹图像,该算法大大提高了方向图准确性,而且自适应Gabor滤波器在处理不同类型的指纹时更具优越性。     

改进的Gabor滤波算法及其在FPGA中的实现

针对指纹图像核心区域方向变化剧烈及用软件实现图像增强速度慢的缺点,提出了一种适应于硬件并行处理的基于改进的Gabor滤波的指纹图像增强算法,并用硬件描述语言在可编程逻辑门阵列(FPGA)上实现,同时介绍了系统的工作过程及总体结构.实验结果表明,改进的Gabor滤波能使图像的断裂处得到很好的连接,指纹图像信息明显增强,并且在速度方面利用硬件实现远远高于软件.实验表明,FPGA在图像处理方面满足实时性的要求,适用于图像增强的系统.     

基于Gabor滤波的指纹图像增强算法

由于指纹的唯一性和不变性,指纹识别已成为当前最流行、最方便、最可靠的个人身份认证技术之一.指纹识别一般包括指纹采集、图像预处理、特征提取及特征匹配等几个步骤.其中图像预处理中的图像增强是最为关键的环节,直接影响特征提取与特征匹配.提出了一种基于Gabor滤波的指纹图像增强算法.阐述了Gabor滤波器的定义及其在指纹图像增强中的应用.对指纹图像方向图提取方法和频率计算方法作了改进尝试.实践表明,该方法是有效的、实用的.     

指纹图像增强和纹线提取算法

研究并实现了指纹图像增强和纹线提取算法。改进了指纹脊线距离求取算法和基于脊线方向的纹线提取算法。实践表明,改进算法工作稳定,效果良好,鲁棒性强,对低质量图像具有显著的增强效果。     

Removal of digitization errors in fingerprint ridgelines using B-splines

Most of the contemporary automatic fingerprint identification systems (AFIS) are based on a dual strategy of combining the minutiae information with the ridge topography in order to improve the overall matching performance. To ensure the efficiency and robustness of such an AFIS, it is necessary, therefore, to rectify the abnormalities or aberrations of the underlying ridge topography, in general, and to smoothen the uneven/noisy ridgelines, in particular. The proposed work deals with one such problem besetting fingerprint analysis—the problem of eliminating digitization errors that usually creep in during fingerprint acquisition or during preprocessing. The method mainly involves fitting of B-splines for a set of control points chosen appropriately for each ridgeline in a fingerprint image. These fitted splines, in turn, can be used to reconstruct the concerned fingerprint, which, after the rectification procedure, becomes almost devoid of such digitization error. With a proper “smoothness parameter” that determines the extent to which a ridgeline is smoothed, the structural information of the corrected ridgelines produces improved results on fingerprint matching. Experimental results on several databases have been reported, which clearly demonstrate the strength and elegance of the proposed algorithm.     

A new model for fingerprint classification by ridge distribution sequences

In this paper, a new method is introduced which is a combination of structural and syntactic approaches for fingerprint classification. The goal of the proposed ridge distribution (R-D) model is to present the idea of the possibility for classifying a fingerprint into the complete seven classes in the Henry's classification. From our observation, there exist only 10 basic ridge patterns which construct fingerprints. Fingerprint classes can be interpreted as a combination of these 10 ridge patterns with different ridge distribution sequences. In this paper, the classification task is performed depending on the global distribution of the 10 basic ridge patterns by analyzing the ridge shapes and the sequence of ridges distribution. The regular expression for each class is formulated and a NFA model is constructed accordingly. An explicit rejection criterion is also defined in this paper. For the seven-class fingerprint classification problem, our method can achieve the classification accuracy of 93.4% with 5.1% rejection rate. For the five-class problem, the accuracy rate of 94.8% is achieved. Experimental results reveal the feasibility and validity of the proposed approach in fingerprint classification.     

一种新的基于距离的汉字笔画抽取方法

在定义了相关术语的基础上提出了一种新的基于距离的汉字笔画抽取方法，并给出了实验结果，结果表明该方法可有效可行，为后续基于笔画的汉字信息处理打下了坚实的基础。     

岭回归法用于五种酚的同时测定

将岭参数引入最小二乘法中,提出了一种新的计算分光光度法－岭回归分光光度法。将该法应用于具紫外吸收的五组分酚类化合物的同时测定。文中详细介绍了该法的基本原理,岭参数以及介质的选择,讨论了波长测定点数及混合标准溶液份数对计算结果的影响。     

Reduced Rank Kernel Ridge Regression

Ridge regression is a classical statistical technique that attempts to address the bias-variance trade-off in the design of linear regression models. A reformulation of ridge regression in dual variables permits a non-linear form of ridge regression via the well-known kernel trick. Unfortunately, unlike support vector regression models, the resulting kernel expansion is typically fully dense. In this paper, we introduce a reduced rank kernel ridge regression (RRKRR) algorithm, capable of generating an optimally sparse kernel expansion that is functionally identical to that resulting from conventional kernel ridge regression (KRR). The proposed method is demonstrated to out-perform an alternative sparse kernel ridge regression algorithm on the Motorcycle and Boston Housing benchmarks.     

基于骨架的血细胞图像分离算法

文章从细胞图像骨架的角度出发,研究和发展了一种新的粘连细胞分离算法。其主要思路是:首先采用细化算法提取细胞图像的骨架,并计算骨架各点的边界距离值,然后利用骨架边界距离函数波谷的特性将相互粘结的细胞分离。新算法提出了一种计算骨架边界距离值的新方法用于分析骨架波谷的特性,并在理论和实践的基础上,研究和发展了细胞图像中粘结细胞的分离准则。为证明该算法的有效性,选取了上百种不同的细胞图像进行了测试。实验结果表明,新算法能有效、快速地对细胞图像中相互粘结的细胞进行合理分割。     

Finding ridges and valleys in a discrete surface using a modified MLS approximation

Implicit surface fitting is a promising approach to finding ridges and valleys in discrete surfaces, but existing techniques are time-consuming and rely on user-supplied tuning parameters. We use a modified MLS (moving-least-squares) approximation technique to estimate the local differential information near a vertex by means of an approximating surface. Ridge and valley vertices are easily detected as zero-crossings, and can then be connected along the direction of principal curvature. Our method, demonstrated on several large meshed models, produces a good fit which leads to improved visualization. It does not oscillate and is quick to compute.     

Finding ridges and valleys in a discrete surface using a modified MLS approximation

Implicit surface fitting is a promising approach to finding ridges and valleys in discrete surfaces, but existing techniques are time-consuming and rely on user-supplied tuning parameters. We use a modified MLS (moving-least-squares) approximation technique to estimate the local differential information near a vertex by means of an approximating surface. Ridge and valley vertices are easily detected as zero-crossings, and can then be connected along the direction of principal curvature. Our method, demonstrated on several large meshed models, produces a good fit which leads to improved visualization. It does not oscillate and is quick to compute.     

一种面向指纹识别的鲁棒脊线跟踪算法

由于传统的指纹识别系统大多要进行二值化和细化过程将消耗大量的计算时间,提出了一种基于脊线跟踪的指纹图像特征点提取算法.在灰度级指纹图像上,沿脊线方向自适应跟踪指纹脊线,直至该条脊线终止或与其他脊线相交,得到一幅细化后的指纹骨架图和附在其上的细节点信息.跟踪过程中,在关键点处进行脊线方向估计和局部滤波,跳跃式地获得脊线骨架点.对于提取到的末端点和交叉点,根据指纹图像的结构特征和统计结果相结合进行去伪后处理.实验证明算法的有效性.