学校代号10532学号.doc

指纹特征信息提取 摘要由于指纹的唯一性和不变性，以及指纹识别技术的可行性和实用性，指纹识别已成为当前最流行、最方便、最可靠的个人身份认证技术之一。尽管在此技术上已有多种成型产品，但因为许多核心技术因商业利益而未经公开，以及社会的发展对系统的性能提出了更高的要求，所以从事该领域研究，仍具有重要的理论意义和实用价值。 本文主要对基于细节特征点的指纹自动识别系统的指纹图像预处理、特征提取方法进行研究。在特征提取与后处理部分，本文通过分析指纹图像中存在的各类噪声的拓扑结构，并结合细化指纹图像中细节特征点固有的分布规律，实现了滤除指纹细节伪特征点的方法。作者在计算机上用VisualC++实现了论文中提到的算法，实验结果表明，本论文中的算法在一定程度上达到了令人满意的图像处理效果，论文的研究既具有一定的理论价值，也具有较强的实用价值。 关键词：指纹识别图像预处理特征提取后处理伪特征滤除 Fingerprint Feature Information Extraction Abstract：Because of the uniqueness and invariability of fingerprint, as well as the feasibility and the usability of the fingerprint recognition technology ,the fingerprint identification is becoming one of the most popular, most convenient and most reliable personal authentication technologies, and many fingerprint products have been offered nowadays. However, most of the key method wouldnt be publicized, while better performance is required. So engaging in the research of this domain, still has the important theoretical meaning and practical value. This paper focuses on the study of the fingerprint images preprocessing and feature extraction approach under the automated fingerprint identification systems, which is based on minutia features. In the preprocessing part, this article proposed a set of fairly effective processing flow; In the aspect of feature extraction and post-processing, a approach to eliminate false features is presented. The elimination algorithm of false features is based on structural imformation of all kinds of noises and the distributing disciplinarian of fingerprint features in thinned fingerprint images. Author practices all algorithms mentioned in this paper with Visual C ++ on the PC computer. The experiment results demonstrate that the methods give a fairly satisfying processing effect of fingerprint images to a certain extent. Key Words: fingerprint identification; preprocessing of fingerprint images; feature extraction; post-processing of fingerprint images; elimination of false features