计算机工程与应用 ›› 2024, Vol. 60 ›› Issue (12): 245-251.DOI: 10.3778/j.issn.1002-8331.2303-0370

• 图形图像处理 • 上一篇    下一篇

一种快速的OCT内外指纹提取方法

朱胜明,浦世亮,王海霞,张怡龙,王映宇,王侃磊   

  1. 1.浙江工业大学 计算机科学与技术学院,杭州 310023
    2.杭州海康威视数字技术股份有限公司,杭州 310051
  • 出版日期:2024-06-15 发布日期:2024-06-14

Fast OCT Based Internal and External Fingerprint Extraction Method

ZHU Shengming, PU Shiliang, WANG  Haixia, ZHANG Yilong, WANG Yingyu, WANG Kanlei   

  1. 1.College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310023, China
    2.Hangzhou Hikvision Digital Technology Co., Ltd., Hangzhou 310051, China
  • Online:2024-06-15 Published:2024-06-14

摘要: 针对现有光学层析成像(optical coherence tomography,OCT)技术中内外部指纹提取算法速度较慢的问题,尤其是许多算法在提取指纹之前需要耗时的原始体数据去噪操作,为了解决这一问题,提出了一种简单快速的OCT内外部指纹提取算法。对于OCT指纹体数据,首先根据每张轴向切片(B-scan)图像的一阶微分最大值,拟合一条基本覆盖角质层轮廓的曲线,然后再根据一阶微分的次最大值,拟合出一条基本穿过活性表皮交界处轮廓脊谷处的曲线,最后在这两个曲线上下各N个像素点的位置,利用两条曲线附近的灰度值均值信息,按B-scan图片顺序进行拼接,生成提取到的内外部指纹图像。通过实验确定了N的最佳取值,然后与现有的内外部指纹提取算法对比,证明所提出的算法能够在不降低指纹质量的情况下,快速地提取出内外部指纹图像,提取内外部指纹的时间比层次聚类算法和梯度法都快很多倍。

关键词: 内部指纹, 外部指纹, 光学层析相干技术, 指纹提取

Abstract: To address the issue of slow internal and external fingerprint extraction algorithms in existing optical coherence tomography (OCT) techniques, especially those requiring time-consuming denoising operations on the raw volumetric data, a simple and fast OCT internal and external fingerprint extraction algorithm has been proposed. For 3D OCT fingerprint data, firstly, according to the maximum first-order differential value of each B-scan image, a curve covering the stratum corneum is fitted. Then, according to the sub-maximum of the first-order differential value ,a curve that basically crosses the ridge and valley of the contour at the junction of the active epidermis is fitted. Finally, at the positions of N pixels above and below the two curves, the gray value mean information near the two curves is used to spliced according to the sequence of B-scan images to generate internal and external fingerprint images. The optimal value of N is determined through experiments. Compared with existing internal and external fingerprint extraction algorithms, it is proved that the proposed algorithm can quickly extract internal and external fingerprint images without reducing fingerprint quality, and the extraction time of internal and external fingerprint is much faster than the hierarchical clustering algorithm and gradient-based methods.

Key words: internal fingerprint, external fingerprint, optical coherence tomography (OCT), fingerprint extraction