毕业设计—指纹图像多级分类算法的研究.docVIP

摘 要 指纹唯一性和终生不变性的特征使其成为理想的身份确认工具。随着计算机与信息处理技术的飞速发展，指纹识别作为是生物特征识别领域中应用最广泛的技术之一，目前普遍应用于身份验证和识别领域，特别是在电子商务、犯罪识别、信息安全等领域广泛的应用。指纹分类为指纹识别提供了捷径，因此对自动指纹分类的研究具有重要的学术价值和实用价值。 指纹分类技术为大型指纹数据库提供了一种重要的索引机制。指纹图象被划分到预先设定的某个类别，这样匹配工作就可在一个子类中进行，从而大大节省运算和匹配时间。本课题在以前分类的基础之上提出了一种新的指纹图像分类方法，即基于模式与结构相结合的方法。先根据有无核心点把指纹图像分成了普通弓型指纹和待识别型指纹，实现一级分类；再基于BP神经网络分类器，把待识别型指纹分为帐弓型、左旋型、右旋型、斗型指纹，实现第二级分类。 基于理论分析的基础上，我们利用MATLAB对整个指纹分类过程进行了编程。 实验结果表明，我们能有效地执行指纹分类，具有一定的实用价值。 关键词：指纹分类、核心点、方向信息、特征提取、BP神经网络。 Abstract Fingerprint recognition technology is one of the most significant and reliable identification methods as it is unique and invariable. With the development of computer science, fingerprint recognition has extensive outlook servicing people. It is widely used in the field of international trade, criminal act and a course of justice. So the research of an automatic fingerprint identification system (AFIS) has important technical value and practical value. Fingerprint classification provides indexing mechanism for a large fingerprint database. Fingerprint image is classified into predefined categories. It can greatly reduce matching time because the matching is implemented in a sub-category. The subject on the basis of the previous classification propose a new fingerprint classification method which is a combination of the model-based and the structure-based. Firstly, we devide the fingerprint into an arch an waiting recognized fingerprint under the core which they exist to achieve a classification. Based on BP Neural Network Classifier ,we devide the waiting recognized finger images into Tented Arch 、Left Loop 、Right Loop and Whorhl, achieving the seond-classification. Based on the theoretical analysis foundation, I used MATLAB to carry on the programming to the entire fingerprint classification process. Experimental results show that we can effectively implement fingerprint classification . They have some practical value . Keywords: Fingerprint classification, C