基于多角度旋转积分图的手背静脉身份识别

随着生物特征识别技术水平的飞速发展,手背静脉识别也广泛运用于各个领域。由于采集终端硬件设备和采集环境的差异,会降低识别的准确性。针对手背静脉图像在亮度,旋转,尺寸等方面造成的影响,,提出了基于多角度旋转积分图的和离散余弦变换的手背静脉图像识别方法,并进行参数优化。本文结合梯度增强的静脉图像分割方法,选取最佳角度间隔做旋转积分运算,然后通过二维离散余弦变换(DCT)截取最佳特征矩阵用做分类识别,识别率超过99.9％。 实验通过对比其它传统算法对手背静脉图像的识别效果来验证本文特征提取方法的可行性和优越性。     

Efficient yet robust biometric iris matching on smart cards for data high security and privacy

Smart control access to any service and/or critical data is at the very basis of any smart project. Biometrics have been used as a solution for system access control, for many years now. However, the simple use of biometrics cannot be considered as final and perfect solution. Most problems are related to the data transmission method between the medias, where the users require access and the servers where the biometric data, captured upon registration, are stored. In this paper, we use smart cards as an effective yet efficient solution to this critical data storage problem. Furthermore, iris texture has been used as a human identifier for some time now. This biometric is considered one of the most reliable to distinguish a person from another as its unique yet perfectly stable over time. In this work, we propose an efficient implementation of iris texture verification on smart cards. For this implementation, the matching is done on-card. Thus, the biometric characteristics are always kept in the owner’s card, guaranteeing the maximum security and privacy. In a first approach, the False Acceptance Rate (FAR) and False Rejection Rate (FRR) are improved using circular translations of the matched iris codes. However, after a thorough analysis of the achieved results, we show that the proposed method introduces a significant increase in terms of execution time of the matching operation. In order to mitigate this impact, we augmented the proposed technique with acceptance threshold verification, thus decreasing drastically the execution time of the matching operation, and yet achieving considerably low FAR and FRR. It is noteworthy to point out that these characteristics are at the basis of any access control successful usage.     

一种基于模板匹配的手指静脉识别算法

相对于指纹识别等传统生物特征识别手段,手指静脉识别是一种新兴的具有较好应用前景的生物特征识别技术。本文设计了具有自适应光源系统的手指静脉采集仪,能够自动获得亮度均匀的手指静脉图像;提出了一种基于模板匹配的手指静脉识别算法,采用基于多方向灰度谷底搜寻方法提取手指静脉特征,然后将从同一手指多个图像中提取的静脉特征合成模板,并通过门限阈值消除模板中的随机差异信息。实验结果表明,运用本研究提出的基于模板匹配的手指静脉识别算法能有效提高识别准确性,具有99.10%的识别准确率和1.03%的等错误率。     

Contact-free hand geometry-based identification system

This paper presents an approach for personal identification using hand geometrical features, in which the infrared illumination device is employed to improve the usability of this hand recognition system. In the proposed system, prospective users can place their hand freely in front of the camera without any pegs or templates. Moreover, the proposed system can be widely used under dark environment and complex background scenarios. To achieve better detection accuracy, in total 13 important points are detected from a palm image, and 34 features calculated from these points are used to further recognition. Experimental results demonstrate that the averaged Correct Identification Rate (CIR) is 96.23% and averaged False Accept Rate (FAR) is 1.85%. These results prove that the proposed contact-free system can be considered as an effective identity verification system for practical applications.     

基于图像采集质量评价的指纹与指静脉双模态识别决策级融合方法

针对单一模态生物特征识别系统性能受图像质量影响较大问题,提出一种基于图像采集质量评价的指纹与指静脉识别的决策级融合方法。该方法不仅对指纹图像进行质量评价,并首次根据指静脉图像特性设计图像采集质量评价指标,以达到克服图像质量对识别结果影响的目的。再针对这两种模态图像特点分别进行分类器设计,得出各自的识别结果后,结合上述得到的图像采集质量评价分数进行决策级融合,将融合后的结果作为最终的识别结果。实验表明,该方法有效克服图像质量对识别结果的影响,提高识别系统的性能,为多生物特征身份识别提供一种有效途径。     

Fusion of palm-phalanges print with palmprint and dorsal hand vein

To ensure the high performance of a biometric system, various unimodal systems are combined to evade their constraints to form a multimodal biometric system. Here, a multimodal personal authentication system using palmprint, dorsal hand vein pattern and a novel biometric modality “palm-phalanges print” is presented. Firstly, we have collected a new anterior hand database of 50 individuals with 500 images at the institute referred to as NSIT Palmprint Database 1.0 by using NSIT palmprint device. Then from these anterior hand images, database for palmprint and palm-phalanges is created. In this biometric system, the individuals do not have to undergo the distress of using two different sensors since the palmprint and palm-phalanges print features can be captured from the same image, using NSIT palmprint device, at the same time. For dorsal hand vein, Bosphorus Hand Vein Database is used because of the stability and uniqueness of hand vein patterns. We propose fusion of three different biometric modalities which includes palmprint (PP), palm-phalanges print (PPP) and dorsal hand vein (DHV) and perform score level fusion of PP-PPP, PP-DHV, PPP-DHV and PP-PPP-DHV strategies. Lastly, we use K-nearest neighbor, support vector machine and random forest to validate the matching stage. The results proved the validity of our proposed modality and show that multimodal fusion has an edge over unimodal fusion.     

A fast feature selection technique in multi modal biometrics using cloud framework

Biometrics refers to the process that uses biological or physiological traits to identify individuals. The progress seen in technology and security has a vital role to play in Biometric recognition which is a reliable technique to validate individuals and their identity. The biometric identification is generally based on either their physical traits or their behavioural traits. The multimodal biometrics makes use of either two or more of the modalities to improve recognition. There are some popular modalities of biometrics that are palm print, finger vein, iris, face or fingerprint recognition. Another important challenge found with multimodal biometric features is the fusion, which could result in a large set of feature vectors. Most biometric systems currently use a single model for user authentication. In this existing work, a modified method of heuristics that is efficiently used to identify an optimal feature set that is based on a wrapper-based feature selection technique. The proposed method of feature selection uses the Ant Colony Optimization (ACO) and the Particle Swarm Optimization (PSO) are used to feature extraction and classification process utilizes the integration of face, and finger print texture patterns. The set of training images is converted to grayscale. The crossover operator is applied to generate multiple samples for each number of images. The wok proposed here is pre-planned for each weight of each biometric modality, which ensures that even if a biometric modality does not exist at the time of verification, a person can be certified to provide calculated weights the threshold value. The proposed method is demonstrated better result for fast feature selection in bio metric image authentication and also gives high effectiveness security.     

A multimodal-Siamese Neural Network (mSNN) for person verification using signatures and EEG

Signatures have long been considered to be one of the most accepted and practical means of user verification, despite being vulnerable to skilled forgers. In contrast, EEG signals have more recently been shown to be more difficult to replicate, and to provide better biometric information in response to known a stimulus. In this paper, we propose combining these two biometric traits using a multimodal Siamese Neural Network (mSNN) for improved user verification. The proposed mSNN network learns discriminative temporal and spatial features from the EEG signals using an EEG encoder and from the offline signatures using an image encoder. Features of the two encoders are fused into a common feature space for further processing. A Siamese network then employs a distance metric based on the similarity and dissimilarity of the input features to produce the verification results. The proposed model is evaluated on a dataset of 70 users, comprised of 1400 unique samples. The novel mSNN model achieves a 98.57% classification accuracy with a 99.29% True Positive Rate (TPR) and False Acceptance Rate (FAR) of 2.14%, outperforming the current state-of-the-art by 12.86% (in absolute terms). This proposed network architecture may also be applicable to the fusion of other neurological data sources to build robust biometric verification or diagnostic systems with limited data size.     

A novel framework for automatic 3D face recognition using quality assessment

The quality of biometric samples plays an important role in biometric authentication systems because it has a direct impact on verification or identification performance. In this paper, we present a novel 3D face recognition system which performs quality assessment on input images prior to recognition. More specifically, a reject option is provided to allow the system operator to eliminate the incoming images of poor quality, e.g. failure acquisition of 3D image, exaggerated facial expressions, etc.. Furthermore, an automated approach for preprocessing is presented to reduce the number of failure cases in that stage. The experimental results show that the 3D face recognition performance is significantly improved by taking the quality of 3D facial images into account. The proposed system achieves the verification rate of 97.09% at the False Acceptance Rate (FAR) of 0.1% on the FRGC v2.0 data set.     

Face Attribute Convolutional Neural Network System for Data Security with Improved Crypto Biometrics

Due to the enormous usage of the internet for transmission of data over a network, security and authenticity become major risks. Major challenges encountered in biometric system are the misuse of enrolled biometric templates stored in database server. To describe these issues various algorithms are implemented to deliver better protection to biometric traits such as physical (Face, fingerprint, Ear etc.) and behavioural (Gesture, Voice, tying etc.) by means of matching and verification process. In this work, biometric security system with fuzzy extractor and convolutional neural networks using face attribute is proposed which provides different choices for supporting cryptographic processes to the confidential data. The proposed system not only offers security but also enhances the system execution by discrepancy conservation of binary templates. Here Face Attribute Convolutional Neural Network (FACNN) is used to generate binary codes from nodal points which act as a key to encrypt and decrypt the entire data for further processing. Implementing Artificial Intelligence (AI) into the proposed system, automatically upgrades and replaces the previously stored biometric template after certain time period to reduce the risk of ageing difference while processing. Binary codes generated from face templates are used not only for cryptographic approach is also used for biometric process of enrolment and verification. Three main face data sets are taken into the evaluation to attain system performance by improving the efficiency of matching performance to verify authenticity. This system enhances the system performance by 8% matching and verification and minimizes the False Acceptance Rate (FAR), False Rejection Rate (FRR) and Equal Error Rate (EER) by 6 times and increases the data privacy through the biometric cryptosystem by 98.2% while compared to other work.     

基于2D-FFT的掌静脉图像Gabor滤波快速增强法

手掌静脉纹识别技术作为新一代高精度的生物特征识别技术,被广泛用于个人身份鉴定领域.有效提取手掌静脉特征对于手掌静脉分类至关重要.然而,由于采集到的手掌静脉图像的质量较差,必须在识别前对手掌静脉图像进行增强.使用二维离散快速傅里叶变换（2D-FFT）代替传统空域卷积滤波,实现Gabor滤波器与原图像的频域卷积滤波.实验结果显示,本文提出的增强方法,相比较传统的自适应直方图均衡化和Retinex算法具有更佳的增强效果,相比于传统Gabor空域卷积滤波具有更低的计算复杂度,更适用于实时系统.     

Liveness detection for dorsal hand vein recognition

As the identification technology is developed day by day, so is the counterfeit, and any accreditation system can be tricked. Therefore, a complete biometric identification system is supposed to distinguish between real and fake. Aiming at the liveness detection problems during the dorsal hand vein (DHV) recognition process, this paper proposes a method which combines principal component analysis and power spectrum estimation of the AR model together, three kinds of fake hand vein images which are paper printed, wearing thin rubber gloves and wearing thick rubber gloves have tested, and the result shows that the recognition rate of fake samples can reach 98.3 %, which proves that this method can realize in liveness detection of DHVs effectively.     

Biometric authentication

In this age of digital impersonation, biometric techniques are being used increasingly as a hedge against identity theft. The premise is that a biometric - a measurable physical characteristic or behavioral trait - is a more reliable indicator of identity than legacy systems such as passwords and PINs. There are three general ways to identify yourself to a computer system, based on what you know, what you have, or who you are. Biometrics belong to the "who you are" class and can be subdivided into behavioral and physiological approaches. Behavioral approaches include signature recognition, voice recognition, keystroke dynamics, and gait analysis. Physiological approaches include fingerprints; iris and retina scans; hand, finger, face, and ear geometry; hand vein and nail bed recognition; DNA; and palm prints. In this article, we focus on the two most popular biometric techniques: fingerprints and iris scans.     

Palm vein recognition using adaptive Gabor filter

Vein pattern recognition is one of the newest biometric techniques researched today. One of the reliable and robust personal identification authentication approaches using palm vein patterns is presented in this paper. In our work, we consider the palm vein as a piece of texture and apply texture-based feature extraction techniques to palm vein authentication. A Gabor filter provides the optimized resolution in both the spatial and frequency domains, thus it is a basis for extracting local features in the palm vein recognition. However, Gabor filter has many potential parameter combinations to use, and it is a common practice now to use multiple Gabor filters or to determine desired single combination by experience. The overall aim of this work is to discuss the optimization algorithm that determines the best parameter values of a single Gabor filter for palm vein recognition. In order to obtain effective pattern of palm vascular, we proposed an innovative and robust adaptive Gabor filter method to encode the palm vein features in bit string representation. The bit string representation, called VeinCode, offers speedy template matching and enables more effective template storage and retrieval. The similarity of two VeinCodes is measured by normalized Hamming distance. A total of 4140 palm vein images were collected form 207 persons to verify the validity of the proposed palm vein recognition approach. High accuracy has been obtained by the proposed method and the speed of this method is rapid enough for real-time palm vein recognition. Experimental results demonstrate that our proposed approach is feasible and effective in palm vein recognition.     

基于多图融合和改进Xception网络的跨设备手背静脉识别研究

手背静脉是一种新兴的生物特征识别技术,相比其他生物特征具有唯一性、防伪造性、稳定性和非接触性等明显优势;由于采集设备和采集环境的不同,手背静脉灰度图像存在亮度、角度旋转、尺度缩放等差异,识别率较低;由此提出一种基于多图融合和Xception网络的手背静脉识别算法;首先在图像预处理后分割得到二值纹理图,然后将二值图转换为距离图,再由二值图细化得到骨架图;最后融合二值图、距离图和骨架图,得到包含纹理特征和形状特征的三通道合并图;采用Xception结构作为分类网络,并将其激活函数ReLU改为非线性更强的h-swish激活函数;相关实验在由实验室自建的1库和2库两个数据库上进行,其中1库作为训练集,2库作为测试集,最高识别率达到93.54％.