基于Android智能手机的心音身份识别系统
发布时间：2018-09-06 07:26
【摘要】：随着信息科技的飞速发展，个人信息的安全性对于人类越来越重要，传统身份识别技术已经不能满足人类对于高安全性身份识别的要求。生物特征识别技术（BiometricsIdentificationTechnology）作为一种利用人体的行为特征、生理特征来进行身份辨别的技术具有更高的安全度与可靠度，已逐渐成为国际研究的热点，并逐渐在不同的领域发挥着举足轻重的作用。常见的生物特征识别技术主要有：虹膜识别、指纹识别、视网膜识别、语音识别、人脸识别以及DNA识别等。这些身份识别技术在实际应用中虽然取得了成功，但是也面临着一些安全性挑战，探索新的生物特征识别方法是生物特征识别技术未来发展方向之一。 本文将来自于人体内部的心音信号用于身份识别，开发了基于Android智能手机的心音身份识别系统。在心音信号的特征提取、模式匹配等关键技术的相关研究基础上，设计了心音信号的硬件采集设备，并在Android智能手机中实现了相应的心音身份识别应用软件。本文主要的研究内容如下： （1）研究了心音身份识别的原理，回顾了心音身份识别技术的国内外研究现状。并介绍了Android系统的体系结构、系统特点、应用组件以及开发环境的构建。 （2）开发了心音信号采集的硬件平台，，主要包括心音传感器、滤波电路、dsPIC主控制电路、增益控制电路、HC-06蓝牙传输电路以及NFC标签等。通过自制心音采集头采集心音，通过滤波电路预处理心音，经过dsPIC主控制电路的A/D转换后，利用NFC标签，通过蓝牙模块完成无线传输。NFC标签的主要作用是简化蓝牙配对的过程，快速建立Android智能手机与心音采集硬件电路的蓝牙连接，实现了用户的便捷操作。 （3）提出了基于MFCC和改进VQ的身份识别算法。首先提出了基于高阶香农熵的心音分段预处理算法，在分段的基础上提取心音的MFCC特征参数；研究了利用VQ模型训练心音码本的原理，给出了利用LBG算法生成码本的方法，最后通过基于方差和标准差的加权失真测度的VQ模型实现了个体身份识别。 （4）基于Android操作系统开发了心音身份识别应用程序，由蓝牙连接、用户注册、用户辨识以及系统管理4个功能模块组成的。在蓝牙连接模块，使用NFC技术简化蓝牙连接的方式。用户注册模块，进行用户基本信息的登记、采集与保存心音信号、心音MFCC特征提取以及基于LBG算法的码本生成。在用户辨识模块，实现身份的辨识。系统管理功能模块的主要功能是注册用户信息的管理。本文提供了一种基于Android的心音身份识别系统，为生物特征识别领域提供了一种安 全可靠便携的身份识别新装备。
[Abstract]:With the rapid development of information technology, the security of personal information is becoming more and more important to human beings. Biometric identification technology (BiometricsIdentificationTechnology), as a technology for identity identification based on human behavior and physiological characteristics, has become a hot topic in international research, because of its higher safety and reliability. And gradually playing a pivotal role in different fields. Common biometric recognition techniques include iris recognition, fingerprint recognition, retinal recognition, speech recognition, face recognition and DNA recognition. Although these technologies have been successful in practical application, they also face some security challenges. Exploring new biometric recognition methods is one of the future development directions of biometric recognition technology. In this paper, a heart sound recognition system based on Android smart phone is developed. On the basis of the research on the key technologies of heart sound signal extraction and pattern matching, the hardware acquisition equipment of heart sound signal is designed, and the corresponding application software of heart sound identity recognition is implemented in Android smart phone. The main contents of this paper are as follows: (1) the principle of heart sound identity recognition is studied, and the research status of heart sound identity recognition technology at home and abroad is reviewed. The architecture, characteristics, application components and development environment of Android system are introduced. (2) the hardware platform of heart sound signal acquisition is developed, which mainly includes heart sound sensor, filter circuit and dsPIC main control circuit. Gain control circuit HC- 06 Bluetooth transmission circuit and NFC label. The heart sound is collected by self-made heart sound acquisition head, the heart sound is preprocessed by filter circuit, after A- / D conversion of dsPIC main control circuit, the main function of wireless transmission. NFC tag is to simplify the process of Bluetooth pairing by using NFC tag and Bluetooth module. The Bluetooth connection between the Android smart phone and the heart sound acquisition hardware circuit is established quickly, and the user's convenient operation is realized. (3) the identification algorithm based on MFCC and improved VQ is proposed. Firstly, a preprocessing algorithm of heart sound segmentation based on high order Shannon entropy is proposed to extract the MFCC characteristic parameters of heart sound on the basis of segmentation, and the principle of training heart sound codebook with VQ model is studied, and the method of generating heart sound codebook using LBG algorithm is given. Finally, the VQ model based on variance and standard deviation is used to realize individual identity recognition. (4) based on Android operating system, a heart sound identification application is developed, which is connected by Bluetooth and registered by users. User identification and system management are composed of four functional modules. In Bluetooth connection module, NFC technology is used to simplify Bluetooth connection mode. In the user registration module, the basic information of the user is registered, the heart sound signal is collected and saved, the heart sound MFCC feature is extracted and the codebook is generated based on the LBG algorithm. In the user identification module, identity identification is realized. The main function of the system management module is the management of registered user information. A heart sound identification system based on Android is proposed in this paper.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN912.34
本文编号：2225662
[Abstract]:With the rapid development of information technology, the security of personal information is becoming more and more important to human beings. Biometric identification technology (BiometricsIdentificationTechnology), as a technology for identity identification based on human behavior and physiological characteristics, has become a hot topic in international research, because of its higher safety and reliability. And gradually playing a pivotal role in different fields. Common biometric recognition techniques include iris recognition, fingerprint recognition, retinal recognition, speech recognition, face recognition and DNA recognition. Although these technologies have been successful in practical application, they also face some security challenges. Exploring new biometric recognition methods is one of the future development directions of biometric recognition technology. In this paper, a heart sound recognition system based on Android smart phone is developed. On the basis of the research on the key technologies of heart sound signal extraction and pattern matching, the hardware acquisition equipment of heart sound signal is designed, and the corresponding application software of heart sound identity recognition is implemented in Android smart phone. The main contents of this paper are as follows: (1) the principle of heart sound identity recognition is studied, and the research status of heart sound identity recognition technology at home and abroad is reviewed. The architecture, characteristics, application components and development environment of Android system are introduced. (2) the hardware platform of heart sound signal acquisition is developed, which mainly includes heart sound sensor, filter circuit and dsPIC main control circuit. Gain control circuit HC- 06 Bluetooth transmission circuit and NFC label. The heart sound is collected by self-made heart sound acquisition head, the heart sound is preprocessed by filter circuit, after A- / D conversion of dsPIC main control circuit, the main function of wireless transmission. NFC tag is to simplify the process of Bluetooth pairing by using NFC tag and Bluetooth module. The Bluetooth connection between the Android smart phone and the heart sound acquisition hardware circuit is established quickly, and the user's convenient operation is realized. (3) the identification algorithm based on MFCC and improved VQ is proposed. Firstly, a preprocessing algorithm of heart sound segmentation based on high order Shannon entropy is proposed to extract the MFCC characteristic parameters of heart sound on the basis of segmentation, and the principle of training heart sound codebook with VQ model is studied, and the method of generating heart sound codebook using LBG algorithm is given. Finally, the VQ model based on variance and standard deviation is used to realize individual identity recognition. (4) based on Android operating system, a heart sound identification application is developed, which is connected by Bluetooth and registered by users. User identification and system management are composed of four functional modules. In Bluetooth connection module, NFC technology is used to simplify Bluetooth connection mode. In the user registration module, the basic information of the user is registered, the heart sound signal is collected and saved, the heart sound MFCC feature is extracted and the codebook is generated based on the LBG algorithm. In the user identification module, identity identification is realized. The main function of the system management module is the management of registered user information. A heart sound identification system based on Android is proposed in this paper.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN912.34
【参考文献】
相关期刊论文 前10条
1 ;虹膜以及视网膜识别技术现状分析[J];中国防伪报道;2006年04期
2 刘磊;周作栋;彭永昱;;安卓手机的前景与现状[J];成功(教育);2013年01期
3 陈洪京;;几种生物识别方法的比较研究[J];河北省科学院学报;2007年04期
4 刘娟;赵治栋;;基于心音信号谱分析的身份特征提取算法[J];杭州电子科技大学学报;2010年04期
5 刘雅琴;裘雪红;;一种加权矢量量化的说话人识别技术[J];计算机与信息技术;2005年06期
6 王新沛;刘常春;李远洋;孙处然;;基于高阶香农熵的心音分段算法[J];吉林大学学报(工学版);2010年05期
7 郭兴明;林辉杰;肖守中;;心音中医学指标的提取[J];计算机工程与应用;2011年03期
8 熊建斌;王钦若;邓九英;刘奇;叶宝玉;;巴氏距离与PCA结合的人脸识别[J];计算机工程与应用;2012年03期
9 田启川;刘正光;;虹膜识别综述[J];计算机应用研究;2008年05期
10 郭兴明;段峗;钟丽莎;;基于HMM和WNN的心音信号身份识别研究[J];计算机应用研究;2010年12期
相关博士学位论文 前1条
1 王守佳;基于图像的人体检测跟踪和人脸识别的研究[D];吉林大学;2013年
本文编号：2225662
本文链接：https://www.wllwen.com/kejilunwen/wltx/2225662.html
