一种基于渐进聚焦的低质量指纹姿态估计方案

doi:10.3969/j.issn.1671-1122.2025.07.013

摘要/Abstract

摘要：

针对指纹姿态估计算法在低质量指纹和扭曲指纹识别中准确率不足的问题，文章提出一种基于ABSF和渐进聚焦的指纹姿态估计方案。该方案首先对原始指纹图像进行预处理获得纹理图像，通过增强算法处理纹理图像得到指纹增强图像，然后基于增强图像预估参考点区域，最后依据质量评估标准判断参考点定位准确性。若未通过判断，则对焦点区域脊线实施质量增强；若通过判断，则执行最终姿态估计。在FVC、NIST SD27和DF 3个标准指纹数据库上的实验结果表明，相较现有方法，该方案展现出更优的姿态估计精度和识别准确率。

关键词: 指纹识别, 姿态估计, 参考点检测

Abstract:

To address the low accuracy issue of fingerprint pose estimation algorithms in processing low-quality and distorted fingerprints, this paper proposed an ABSF-based progressive focusing scheme. The workflow included: preprocessing raw fingerprint images to obtain texture images; generating enhanced fingerprint images through image enhancement; estimating reference point regions from enhanced images; evaluating positioning accuracy using quality criteria-performing ridge quality enhancement for unqualified cases, or executing final pose estimation for qualified cases. Experimental validation on FVC, NIST SD27 and DF benchmark databases demonstrates superior estimation precision and recognition accuracy compared to existing methods.

Key words: fingerprint recognition, attitude estimation, reference point detection

中图分类号:

TP309

张雪锋, 苗锴. 一种基于渐进聚焦的低质量指纹姿态估计方案[J]. 信息网络安全, 2025, 25(7): 1153-1162.

ZHANG Xuefeng, MIAO Kai. A Progressive Focusing-Based Scheme for Low-Quality Fingerprint Pose Estimation[J]. Netinfo Security, 2025, 25(7): 1153-1162.

图/表 20

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

表1

图15

表2

图16

图17

表3

参考文献 25

[1]	TAIGMAN Y, YANG Ming, RANZATO M, et al. DeepFace: Closing the Gap to Human-Level Performance in Face Verification[C]// IEEE. The IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2014: 1701-1708.
[2]	MALTONI D, MAIO D, JAIN A K, et al. Handbook of Fingerprint Recognition[M]. Heidelberg: Springer, 2022.
[3]	DUAN Yongjie, FENG Jianjiang, LU Jiwen, et al. Estimating Fingerprint Pose via Dense Voting[J]. Transactions on Information Forensics and Security, 2023, 18: 2493-2507.
[4]	SU Yijing, FENG Jianjiang, ZHOU Jie. Fingerprint Indexing with Pose Constraint[J]. Pattern Recognition, 2016, 54: 1-13.
[5]	YIN Qihao, FENG Jianjiang, LU Jiwen, et al. Orientation Field Estimation for Latent Fingerprints by Exhaustive Search of Large Database[C]// IEEE. The 9th International Conference on Biometrics Theory, Applications and Systems (BTAS). New York: IEEE, 2018: 1-9.
[6]	OUYANG Jiahong, FENG Jianjiang, LU Jiwen, et al. Fingerprint Pose Estimation Based on Faster R-CNN[C]// IEEE. International Joint Conference on Biometrics (IJCB). New York: IEEE, 2017: 268-276.
[7]	YANG Xiao, FENG Jianjiang, ZHOU Jie. Localized Dictionaries Based Orientation Field Estimation for Latent Fingerprints[J]. Pattern Analysis and Machine Intelligence, 2014, 36(5): 955-969.
[8]	GU Shan, FENG Jianjiang, LU Jiwen, et al. Efficient Rectification of Distorted Fingerprints[J]. Information Forensics and Security, 2017, 13(1): 156-169.
[9]	LIU Manhua, JIANG Xudong, KOT A C. Fingerprint Reference-Point Detection[J]. Journal on Advances in Signal Processing, 2005(4): 498-509.
[10]	DEERADA C, PHROMSUTHIRAK K, RUNGCHOKANUN A, et al. Progressive Focusing Algorithm for Reliable Pose Estimation of Latent Fingerprints[J]. Information Forensics and Security, 2019, 15: 1232-1247.
[11]	RERKRAI K, AREEKUL V. A New Reference Point for Fingerprint Recognition[C]// IEEE. International Conference on Image Processing. New York: IEEE, 2000: 499-502.
[12]	SUTTHIWICHAIPORN P, AREEKUL V. Adaptive Boosted Spectral Filtering for Progressive Fingerprint Enhancement[J]. Pattern Recognition, 2013, 46(9): 2465-2486.
[13]	FIELD D J. Relations Between the Statistics of Natural Images and the Response Properties of Cortical Cells[J]. Journal of the Optical Society of America A-Optics, 1987, 4(12): 2379-2394.
[14]	CHAN S H, KHOSHABEH R, GIBSON K B, et al. An Augmented Lagrangian Method for Total Variation Video Restoration[J]. Image Processing, 2011, 20(11): 3097-3111.
[15]	CHIKKERUR S, CARTWRIGHT A N, GOVINDARAJU V. Fingerprint Enhancement Using STFT Analysis[J]. Pattern Recognition, 2007, 40(1): 198-211.
[16]	LI Zeqi. Study on the Low-Quality Fingerprint Image Enhancement Based on STFT[D]. Harbin: Harbin University of Science and Technology, 2023.
	李泽奇. 基于STFT的低质量指纹图像增强算法研究[D]. 哈尔滨: 哈尔滨理工大学, 2023.
[17]	MENG Zhi. Research on Low Quality Fingerprint Matching Algorithm Based on Small Area Phase Correlation[D]. Kunming: Faculty of Science Kunming University of Science and Technology, 2023.
	孟志. 基于小面积相位相关低质量指纹匹配算法的研究[D]. 昆明: 昆明理工大学, 2023.
[18]	WANG Juan. Research on the Low-Quality Fingerprint Image Enhancement Algorithm Based on Signal Analysis[D]. Harbin: Harbin University of Science and Technology, 2023.
	王娟. 基于信号解析的低质量指纹图像增强算法研究[D]. 哈尔滨: 哈尔滨理工大学, 2023.
[19]	UMBAUG S E. Digital Image Processing and Analysis: Applications with MATLAB and CVIPtools[M]. Boca Raton: CRC Press, 2017.
[20]	BAZEN A M, GEREZ S H. Systematic Methods for the Computation of the Directional Fields and Singular Points of Fingerprints[J]. Pattern Analysis and Machine Intelligence, 2002, 24(7): 905-919.
[21]	LLOYD S. Least Squares Quantization in PCM[J]. Information Theory, 1982, 28(2): 129-137.
[22]	SI Xuanbin, FENG Jianjiang, ZHOU Jie, et al. Detection and Rectification of Distorted Fingerprints[J]. Pattern Analysis and Machine Intelligence, 2015, 37(3): 555-568.
[23]	VeriFinger SDK. Fingerprint Identification for Stand-Alone or Client-Server Solutions[EB/OL]. (2023-04-18) [2024-11-22]. https://www.neurotechnology.com/verifinger.html.
[24]	ZHU En, GUO Xifeng, YIN Jianping. Walking to Singular Points of Fingerprints[J]. Pattern Recognition, 2016, 56: 116-128.
[25]	YIN Qihao, FENG Jianjiang, LU Jiwen, et al. Joint Estimation of Pose and Singular Points of Fingerprints[J]. Information Forensics and Security, 2020(11): 1-3.

窗口尺寸	FVC2000				FVC2002
窗口尺寸	DB1	DB2	DB3	DB4	DB1	DB2	DB3	DB4
16$\times $16	—	√	—	—	√	—	√	—
32$\times $32	√	—	√	√	—	√	—	√
窗口尺寸	FVC2004				FVC2006		DF
窗口尺寸	DB1	DB2	DB3	DB4	DB2	DB3	DB4
16$\times $16	—	—	—	—	—	—	—	—
32$\times $32	√	√	√	√	√	√	√	√

方法	姿态定位误差		未检测到	方向误差
方法	误差平均值	标准差	未检测到	误差平均值	标准差
本文方案	18.3%	19.7%	0	-1.4?	26.7?
文献[10]方案	29.9%	44.8%	0	-2.0?	27.3?
文献[23]方案	83.0%	65.5%	0	-7.0?	89.3?
文献[24]方案	144.2%	81.4%	141	N/A	N/A
文献[25]方案	30.4%	78.9%	0	-1.7?	33.8?

指标	本文方案	移除步骤A	移除步骤B	移除步骤C	移除步骤E
平均误差	28.79%	31.54%	31.62%	30.25%	41.05%
标准差	45.78%	48.97%	43.76%	49.14%	55.25%