非接触式3D自动指纹识别系统

Fingerprint-based biometric is the most proven technique and has the largest market shares. Current automated fingerprint recognition systems usually adopt the contact acquisition mode, and 2D fingerprint features are extracted and used for authentication. The performance (e.g. recognition accuracy) is still much lower than the expectations of people and theory estimation. Many new requirements are also raised along with more and more adoption of fingerprint technique in civilian applications, such as template security, hygiene, user-friendly and so on. For the purpose of further meeting people’s needs (e.g. recognition accuracy, template security, and hygiene etc.), this project proposes a contactless 3D Automated Fingerprint Recognition System (AFRS). Through collecting fingerprint images in a contactless manner, the system successfully resolved the template security issue caused by fingerprints left behind, as well as the hygiene problem due to the need of contact between the human finger and the acquisition sensor. Meanwhile, 3D fingerprint images are generated by a newly designed 3D reconstruction technique specific to fingerprint images. It helps to expand the current fingerprint 2D features into 3D space, and makes 3D fingerprint feature extraction possible. Finally, the recognition accuracy of the proposed 3D AFRS is improved by matching 3D fingerprint features, and fusing with the existing 2D fingerprint features. The project is expected to achieve 3D fingerprint recognition technology, so as to promote the development of new generation fingerprint recognition techniques in our country.

指纹识别是目前最成熟的生物特征识别技术，并拥有最大的市场份额。现有的自动指纹识别系统一般是基于接触式采集方式，并提取指纹的2D特征来进行识别的，其性能（如识别精度）仍然比人的预期和理论估计要低得多。同时，许多新的要求也随着自动指纹识别系统的广泛使用而出现，如指纹模板的安全，卫生，系统使用方便性等问题。为进一步满足人们的需求，本项目提出一种非接触式的3D自动指纹识别系统。该系统通过非接触式的采集方式，成功解决了接触式采集因遗留指纹而造成的模板安全问题，以及因需要手指与采集传感器接触而存在的卫生问题。同时，通过设计针对于指纹图像的3D重构技术，生成3D指纹图像，将目前指纹2D特征扩展到3D空间，使得3D指纹特征提取成为可能。最终，通过提取及匹配3D指纹特征，并与现有的2D指纹特征进行融合，达到提高识别精度的目的。本项目预期实现3D指纹识别技术，促进我国新一代指纹识别核心技术的发展。

当前自动指纹识别系统一般是基于接触式采集方式，并提取指纹的2D特征来进行识别，其性能（如识别精度）仍然比人的预期和理论估计要低得多。同时，许多新的要求也随着自动指纹识别系统的广泛使用而出现，如指纹模板的安全，卫生，系统使用方便性等问题。为进一步满足市场需求，本项目提出一种非接触式3D自动指纹识别系统。通过三年的工作，本项目研究主要完成：(1) 设计了非接触式的采集设备，成功解决了接触式采集方式因遗留指纹而造成的模板安全问题，以及因需要手指与采集传感器接触而存在的卫生问题，采集了一个包括多幅非接触式指纹图像的研究数据库。(2) 通过多目采集方式，生成3D指纹图像，将目前指纹2D特征扩展到3D空间，使得3D指纹特征提取成为可能。(3) 分析3D指纹数据，提出3D特征提取及匹配算法，并与现有的2D指纹特征进行融合，达到高精度识别率的目的。(4) 最终实现一套非接触式指纹识别系统原型，初步实现从采集到识别的完整系统。同时，在本项目的支持下，还进行不局限于指纹识别的算法研究及在其他领域的应用。本项目的研究成果也获得了国内外同行的认可，在项目期间发表学术论文10余篇，其中SCI检索期刊8篇，EI收录会议论文4篇，项目组成员申请发明专利1项。总之，本项目的开展，提出并成功研发一种非接触式3D自动指纹识别系统原型，弥补目前指纹识别系统的不足，为自动指纹识别系统的研究发展提供新思路和可行的解决方案，开拓指纹识别系统研究新领域。