Abstract: A method to extract fingerprint minutiaes was proposed. Acquired binary image, we thinned the ridges and valleys respectively,then extracted ridge bifurcations and valley bifurcations as the key minutiaes from the ridge thinned image and valley thinned image. These minutiaes are more exact and the spurious bifurcations can be removed easily. Experimental results show that the amount of the key minutiaes is less,and it is helpful to fingerprint matching phase.

Key words: direction image, binarization, minutiae extraction

摘要： 提出一种指纹细节点提取方法,在得到指纹二值化图像后,分别对脊线和谷线进行细化,从 脊线细化图和谷线细化图中提取分支点作为指纹的特征点。这样得到指纹的细节点比较准确,而虚假 点也比较容易除去。实验证明,这种方法得到细节点的数目也相对较少,有利于进一步的指纹匹配。

关键词: 方向图, 二值化, 特征点提取