使用支持向量机分割指纹图像的方法

Fingerprint Image Segmentation Based on Support Vector Machines

提出了一种使用支持向量机的指纹图像背景分割方法.以指纹图像的灰度方差和对比度为特征,训练用于分割背景的支持向量机,并用拟凸算法修正支持向量机的分割结果;使用较少的训练样本,得到泛化性能较好的分类器,对较大数据库中的指纹图像分割背景,正确率为95.3%,实现了利用小样本训练分割方法而较好地适用于同类大数据库的目的.

We present a fingerprint image segmentation algorithm based on support vector machines （SVM）. The algorithm uses the features of the gray variance and contrast to train the SVM for image segmentation, and fits the segmentation result of the SVM to convex hull. The experimental results show the algorithm is robust, especially for training segmentation algorithm with a small quantity of samples.