基于手背静脉的身份识别技术研究

发布时间：2018-04-10 22:42

本文选题：手背静脉识别 + 图像预处理　；参考：《哈尔滨理工大学》2017年硕士论文

【摘要】：伴随着信息技术的飞速发展和人们社会活动、经济活动的增加,身份识别变得更加的重要。传统的身份识别(口令识别、密码验证等)方式无法识别用户本身,其安全性较差,不能满足当今社会对身份识别的需要。手背静脉识别技术作为生物特征识别技术之一,为解决这类安全问题带来了希望。本文采用机器视觉识别技术对手背静脉图像进行采集,在此基础上对手背静脉图像进行预处理以及对识别方法的研究。本文设计了基于机器视觉的手背静脉图像采集系统。为了获得高质量、受外界干扰较少的手背静脉图像,本文选择了合适的光源、挡光板、图像传感器以及镜头等部件。为了有效地提取出手背静脉结构的数据特征,需要对图像进行必要的预处理,包括手背摆放的矫正、图像分割、噪声处理、图像增强、血管脉络结构的细化等,同时对图像有效区域进行提取以及图像的归一化处理。本文根据手背血管脉络的特征,采用多种预处理方式,并结合MATLAB仿真得到相应的实验结果,为后面的特征提取和识别奠定了良好的基础。在完成图像预处理的基础上,进行手背静脉特征的提取。针对手背静脉识别的检测要求,建立手背静脉特征提取算法与BP神经网络算法相结合的识别模型。在手背静脉特征提取算法方面,研究了主成份分析-线性判别分析(PCA-LDA)和核二维主成分分析-二维线性判别分析(K2DPCA-2DLDA)方法进行手背静脉特征的提取,并将提取的成分变量作为BP神经网络的输入进行训练、固定网络参数。经过对神经网络进行改进,最后将训练完成的神经网络用于手背静脉的识别。采用MATLAB软件对手背静脉特征提取算法和识别算法进行仿真,仿真结果表明,基于K2DPCA-2DLDA-BP模型的算法在精确度上与实时性方面都有较大提高,达到了研究的目的。将本文研究的手背静脉识别技术在测试系统中应用,通过对测试系统的软件设计,实现了基于手背静脉的身份识别,验证了本论文研究的有效性和实用性。
[Abstract]:With the rapid development of information technology and people's social activities, the increase of economic activities, identity has become more important.Traditional identification (password identification, password authentication) can not identify the user itself, its security is poor, can not meet the needs of today's society for identity recognition.As one of biometric recognition techniques, dorsal hand vein recognition technology brings hope to solve this kind of safety problems.In this paper, the image of the dorsal hand vein is collected by the machine vision recognition technology, and the image of the dorsal hand vein is preprocessed and the method of recognition is studied.In this paper, the image acquisition system of dorsal hand vein based on machine vision is designed.In order to obtain high quality and less interference image of dorsal hand vein, suitable light source, blocking plate, image sensor and lens are selected in this paper.In order to extract the data features of the dorsal hand vein structure effectively, it is necessary to pre-process the image, including the correction of the back of the hand, image segmentation, noise processing, image enhancement, thinning of the vascular structure, etc.At the same time, the effective region of the image is extracted and the image is normalized.In this paper, according to the characteristics of the veins of the back of the hand, various preprocessing methods are adopted and the corresponding experimental results are obtained by MATLAB simulation, which lays a good foundation for the feature extraction and recognition of the back.On the basis of image preprocessing, the feature extraction of dorsal hand vein was carried out.According to the detection requirements of dorsal hand vein recognition, a recognition model combining back hand vein feature extraction algorithm and BP neural network algorithm is established.In the aspect of dorsal hand vein feature extraction, the methods of principal component analysis-linear discriminant analysis (PCA-LDA) and nuclear two-dimensional principal component analysis-2-D linear discriminant analysis (K2DPCA-2DLDAA) were studied to extract the features of dorsal hand vein.The extracted component variables are trained as the input of BP neural network and the network parameters are fixed.After improving the neural network, the trained neural network is applied to the recognition of the dorsal hand vein.The simulation results show that the algorithm based on K2DPCA-2DLDA-BP model can improve the accuracy and real-time performance of the algorithm and achieve the purpose of the research.This paper applies the technology of hand back vein recognition to the testing system. Through the software design of the test system, the identification based on the hand back vein is realized, which verifies the validity and practicability of this paper.
【学位授予单位】：哈尔滨理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【相似文献】