基于方向图的重叠指纹图像分离算法研究
发布时间：2018-08-30 12:40
【摘要】：自动指纹识别系统中指纹自动匹配的一个关键步骤是自动地可靠地从输入的指纹图像中提取细节。然而，细节提取算法的性能好坏与输入图像质量有着密切的关系。为了保证自动指纹识别系统的性能对输入图像的质量的依赖尽量少，确保细节提取算法对图像的质量具有足够的鲁棒性，所以在指纹细节提取算法中结合指纹增强算法是必须的。在通常指纹增强算法中，基本是对单个指纹图像进行降噪工作，而本文提出了对重叠的指纹进行分离的工作，所基于的原理是将其中的一个指纹作为另外一个指纹的噪声，在确定了指纹图像的方向图后，，应用Gabor滤波器作为带通滤波器进行指纹图像增强，将噪声指纹去除，从而达到了将重叠的两个指纹分开的目的。实验表明，本算法能够将重叠的两个指纹分开，具有良好的应用价值。
[Abstract]:A key step of automatic fingerprint matching in automatic fingerprint identification system is to extract details from the input fingerprint image automatically and reliably. However, the performance of the detail extraction algorithm is closely related to the quality of the input image. To ensure the robustness of the detail extraction algorithm to the image quality, it is necessary to combine the fingerprint enhancement algorithm with the fingerprint detail extraction algorithm. One fingerprint is taken as the noise of the other fingerprint. After determining the directional pattern of the fingerprint image, Gabor filter is used as a band-pass filter to enhance the fingerprint image and remove the noisy fingerprint, thus the purpose of separating the overlapped two fingerprints is achieved. Experiments show that the algorithm can separate the overlapped two fingerprints. It has good application value.
【学位授予单位】：辽宁师范大学
【学位级别】：硕士
【学位授予年份】：2002
【分类号】：D918
本文编号：2213094
[Abstract]:A key step of automatic fingerprint matching in automatic fingerprint identification system is to extract details from the input fingerprint image automatically and reliably. However, the performance of the detail extraction algorithm is closely related to the quality of the input image. To ensure the robustness of the detail extraction algorithm to the image quality, it is necessary to combine the fingerprint enhancement algorithm with the fingerprint detail extraction algorithm. One fingerprint is taken as the noise of the other fingerprint. After determining the directional pattern of the fingerprint image, Gabor filter is used as a band-pass filter to enhance the fingerprint image and remove the noisy fingerprint, thus the purpose of separating the overlapped two fingerprints is achieved. Experiments show that the algorithm can separate the overlapped two fingerprints. It has good application value.
【学位授予单位】：辽宁师范大学
【学位级别】：硕士
【学位授予年份】：2002
【分类号】：D918
【参考文献】
相关期刊论文 前7条
1 吴高洪,章毓晋,林行刚;分割双纹理图像的最佳Gabor滤波器设计方法[J];电子学报;2001年01期
2 曾洪波,汪国有,张天序;基于连续方向图的指纹智能预处理算法[J];红外与激光工程;2001年06期
3 邬永革,丁震,黄炯,杨静宇;一种有效的图象特征抽取和识别方法[J];计算机研究与发展;1997年01期
4 赵亚湘;指纹图像预处理方法的研究[J];微机发展;1996年04期
5 梁栋,洪力奋;一种指纹图象的特征提取方法[J];微电子学与计算机;1997年01期
6 张树人,周毅,林泳;指纹图象动识别前的预处理[J];现代计算机;1996年04期
7 桑爱军,王世刚,陈绵书;指纹预处理中的一种新方法[J];仪器仪表学报;2001年S1期
本文编号：2213094
本文链接：https://www.wllwen.com/shekelunwen/gongan/2213094.html
