摘要
指纹图像分割是自动指纹识别系统的重要部分,针对已有指纹分割方法的不足,笔者提出了基于Adaboost的指纹图像分割方法.并且基于分级处理的思想,使用Otsu计算分割阈值,对图像进行初分割,然后又使用笔者提出的基于Adaboost的指纹分割方法对图像进一步的分割,最后采用形态学方法对细分割后的结果进行后处理实验证明,相对于已有的指纹图像分割方法,该方法具有较高的准确性和较强的适应性.
Fingerprint image segmentation is an important part of fingerprint recognition.This paper proposed novel fingerprint image segmentation method based on Adaboost to overcome some shortages in the present literatures.We first segmented fingerprint image with Otsu,and then did the second segmentation with adaboost based on the subsequent processing,and finally morphology was applied as post processing to reduce misclassification.All experimental results show that the method holds superior accuracy and higher adaptability compared with other existing fingerprint image segmentation method.
出处
《山东师范大学学报（自然科学版）》
CAS
2013年第4期24-27,共4页
Journal of Shandong Normal University(Natural Science)
