Ultrasonic Guided Wave Inversion Based on Deep Learning Restoration for Fingerprint Recognition

被引:7
作者
Zhao, Chengwei [1 ]
Li, Jian [1 ]
Lin, Min [1 ,2 ]
Chen, Xin [3 ]
Liu, Yang [1 ,4 ]
机构
[1] Tianjin Univ, State Key Lab Precis Measuring Technol & Instrume, Tianjin 300072, Peoples R China
[2] Univ Wyoming, Dept Mech Engn, Laramie, WY 82071 USA
[3] Southwest Res Inst, Dept Mech Engn, San Antonio, TX 78238 USA
[4] Tianjin Univ, Int Inst Innovat Design & Intelligent Mfg, Shaoxing 312077, Zhejiang, Peoples R China
基金
美国国家科学基金会;
关键词
Fingerprint recognition; Image reconstruction; Tomography; Acoustics; Transmission line matrix methods; Sensors; Noise reduction; Fast inversion tomography (FIT); fingerprint recognition; imaging restoration; ultrasonic guided waves; SUPERVISED DESCENT METHOD; FORM INVERSION; RESOLUTION; REPRESENTATION; TOMOGRAPHY; SYSTEM; CNN; SH;
D O I
10.1109/TUFFC.2022.3198503
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
As an established biometric authentication approach, fingerprint scanning has received considerable attention due to its high accuracy and reliability. In this article, the fingerprint reconstruction at any position is achieved in large physical domains, which monitors wavefield variations of plate-like structures within arrays through the ultrasonic guided wave. Accurate reconstruction and quantitative characterization of fingerprints are obtained using fast inversion tomography (FIT) based on the deep learning convolutional neural network (DLCNN). Parametric optimization is conducted to reveal submillimeter fingerprint minutiae, and a specific DLCNN model is proposed for the artifact removal in FIT reconstructions. The results prove that the FIT based on DLCNN restoration can significantly improve the imaging quality in terms of increased resolution, reduced reconstruction errors, and higher fingerprint matching confidence. The reconstruction also allows an exponential improvement in computational efficiency as a result of much-reduced sensor numbers. Several factors affecting the performance of the proposed reconstruction method are discussed at the end.
引用
收藏
页码:2965 / 2974
页数:10
相关论文
共 45 条
[1]  
Abraham Joshua, 2011, State of the Art in Biometrics, P25
[2]   Trainable Nonlinear Reaction Diffusion: A Flexible Framework for Fast and Effective Image Restoration [J].
Chen, Yunjin ;
Pock, Thomas .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2017, 39 (06) :1256-1272
[3]   Fingerprint Imaging System Based on Capacitive Micromachined Ultrasonic Transducer by Using Impediography Method Including Direct Touch and Waveguide Methods [J].
Choi, Won Young ;
Kwak, Yun Sang ;
Park, Kwan Kyu .
IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL, 2019, 66 (02) :402-411
[4]  
Deng J, 2009, PROC CVPR IEEE, P248, DOI 10.1109/CVPRW.2009.5206848
[5]   Full waveform inversion of SH- and Love-wave data in near-surface prospecting [J].
Dokter, E. ;
Koehn, D. ;
Wilken, D. ;
De Nil, D. ;
Rabbel, W. .
GEOPHYSICAL PROSPECTING, 2017, 65 :216-236
[6]   Nonlocally Centralized Sparse Representation for Image Restoration [J].
Dong, Weisheng ;
Zhang, Lei ;
Shi, Guangming ;
Li, Xin .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2013, 22 (04) :1618-1628
[7]   Fingerprint minutiae extraction from skeletonized binary images [J].
Farina, A ;
Kovács-Vajna, ZM ;
Leone, A .
PATTERN RECOGNITION, 1999, 32 (05) :877-889
[8]   NONLINEAR IMAGE RECOVERY WITH HALF-QUADRATIC REGULARIZATION [J].
GEMAN, D ;
YANG, CD .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 1995, 4 (07) :932-946
[9]   Weighted Nuclear Norm Minimization and Its Applications to Low Level Vision [J].
Gu, Shuhang ;
Xie, Qi ;
Meng, Deyu ;
Zuo, Wangmeng ;
Feng, Xiangchu ;
Zhang, Lei .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2017, 121 (02) :183-208
[10]   Full-waveform inversion imaging of the human brain [J].
Guasch, Lluis ;
Calderon Agudo, Oscar ;
Tang, Meng-Xing ;
Nachev, Parashkev ;
Warner, Michael .
NPJ DIGITAL MEDICINE, 2020, 3 (01)