IP Library › Granted Patent US 12,657,949
Granted Patent B2
US 12,657,949 · App. 18/451,515 · Granted Jun 16, 2026

Method and apparatus for generating palmprint sample, device, and medium

Inventors: Yingyi Zhang (Shenzhen, CN); Kai Zhao (Shenzhen, CN); Lei Shen (Shenzhen, CN); Ruixin Zhang (Shenzhen, CN); Chuhan Zhou (Shenzhen, CN); Tao Wang (Shenzhen, CN); Shouhong Ding (Shenzhen, CN)
Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
G06V40/1347G06V10/422G06V10/44G06V10/774
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,657,949
App. No.
18/451,515
Granted
Jun 16, 2026
Kind
B2
Abstract

A method for generating a palmprint sample includes: generating positioning point data according to a palmprint principal line distribution rule, the positioning point data comprising first data corresponding to a first principal line positioning point and second data corresponding to a second principal line positioning point; generating adjusting point data according to a palmprint principal line curving rule, a principal line adjusting point corresponding to the adjusting point data being configured to control a curving parameter of a principal line formed by the first principal line positioning point and the second principal line positioning point; generating a palmprint principal line based on the first data, the second data and the adjusting point data; and generating at least one palmprint sample comprising the palmprint principal line.

Claims (77)

1 . A method for generating a palmprint sample, executed by a computer device, comprising:

generating positioning point data according to a palmprint principal line distribution rule, the positioning point data comprising first data corresponding to a first principal line positioning point and second data corresponding to a second principal line positioning point;

generating adjusting point data according to a palmprint principal line curving rule, a principal line adjusting point corresponding to the adjusting point data being configured to control a curving parameter of a principal line formed by the first principal line positioning point and the second principal line positioning point;

generating a palmprint principal line based on the first data, the second data and the adjusting point data; and

generating at least one palmprint sample comprising the palmprint principal line.

2 . The method according to claim 1 , wherein the generating positioning point data according to a palmprint principal line distribution rule comprises:

determining a first region and a second region corresponding to a trend of the palmprint principal line according to the palmprint principal line distribution rule;

determining the first data corresponding to the first principal line positioning point in the first region; and

determining the second data corresponding to the second principal line positioning point in the second region.

3 . The method according to claim 2 , wherein the determining a first region and a second region corresponding to a trend of the palmprint principal line according to the palmprint principal line distribution rule comprises:

determining a palmprint generation region according to the palmprint principal line distribution rule, the palmprint generation region indicating a distribution range of the palmprint principal line;

determining a first vertex and a second vertex in a diagonal relationship in the palmprint generation region;

determining the first region in the palmprint generation region by using the first vertex as a center and a first preset length as a radius; and

determining the second region in the palmprint generation region by using the second vertex as a center and a second preset length as a radius.

4 . The method according to claim 2 , wherein the determining the first data corresponding to the first principal line positioning point in the first region and determining the second data corresponding to the second principal line positioning point in the second region comprises:

determining the first data by randomly selecting the first principal line positioning point in the first region; and

determining the second data by randomly selecting the second principal line positioning point in the second region.

5 . The method according to claim 1 , wherein the generating adjusting point data according to a palmprint principal line curving rule comprises:

determining a third region based on a position relationship between the first principal line positioning point and the second principal line positioning point according to the palmprint principal line curving rule; and

generating the adjusting point data in the third region.

6 . The method according to claim 5 , wherein the determining a third region based on a position relationship between the first principal line positioning point and the second principal line positioning point comprises:

connecting the first principal line positioning point and the second principal line positioning point to obtain a target line segment; and

using a rectangular region having a preset edge length as the third region by using a line segment midpoint of the target line segment as a center.

7 . The method according to claim 1 , wherein the palmprint sample further comprises palmprint fine lines; and

the generating at least one palmprint sample comprising the palmprint principal line comprises:

determining at least two palmprint fine line positioning points;

generating the palmprint fine lines based on the at least two palmprint fine line positioning points; and

generating the at least one palmprint sample comprising the palmprint principal line and the palmprint fine lines in a preset palmprint quantity range, the palmprint quantity range comprising at least one of a palmprint principal line quantity range and a palmprint fine line quantity range.

8 . The method according to claim 7 , wherein the generating the palmprint fine lines based on the at least two palmprint fine line positioning points comprises:

connecting the at least two palmprint fine line positioning points to obtain the palmprint fine lines.

9 . The method according to claim 7 , wherein the generating the palmprint fine lines based on the at least two palmprint fine line positioning points comprises:

determining a palmprint fine line adjusting point based on the at least two palmprint fine line positioning points, the palmprint fine line adjusting point being configured to control the curving parameter between the at least two palmprint fine line positioning points; and

determining the palmprint fine lines based on the at least two palmprint fine line positioning points and the palmprint fine line adjusting point in the preset palmprint fine line quantity range.

10 . The method according to claim 1 , further comprising:

acquiring a sample image set, the sample image set storing at least one sample image;

combining the palmprint sample with the sample image by using the sample image as a background to obtain a target image; and

training a palmprint recognition model by using the target image.

11 . The method according to claim 10 , wherein the combining the palmprint sample with the sample image by using the sample image as a background to obtain a target image comprises:

disturbing the at least one palmprint sample in a target disturbing interval to obtain a target sample; and

combining the target sample with the sample image by using the sample image as a background to obtain the target image.

12 . The method according to claim 11 , wherein the target disturbing interval comprises a palmprint principal line disturbing interval and a palmprint fine line disturbing interval;

the disturbing the at least one palmprint sample in a target disturbing interval to obtain a target sample comprises:

disturbing a palmprint principal line corresponding to the at least one palmprint sample in the palmprint principal line disturbing interval to obtain a disturbed principal line;

disturbing palmprint fine lines corresponding to the at least one palmprint sample in the palmprint fine line disturbing interval to obtain disturbed fine lines; and

obtaining the target sample based on the disturbed principal line and the disturbed fine lines.

13 . The method according to claim 11 , wherein the disturbing the at least one palmprint sample comprises:

increasing noise for the palmprint principal line corresponding to the at least one palmprint sample.

14 . The method according to claim 11 , wherein the disturbing the at least one palmprint sample comprises:

increasing noise for the palmprint fine lines corresponding to the at least one palmprint sample.

15 . The method according to claim 10 , wherein the training the palmprint recognition model by using the target image comprises:

performing first training on the palmprint recognition model by using the target image to obtain a candidate palmprint recognition model;

acquiring a palmprint dataset, the palmprint dataset storing at least one palmprint datum, the at least one palmprint datum being correspondingly marked with a data label; and

performing second training on the candidate palmprint recognition model by using the palmprint data and a data label corresponding to the palmprint data to obtain a target palmprint recognition model, the target palmprint recognition model being a model obtained through training the palmprint recognition model.

16 . The method according to claim 15 , wherein the performing second training on the candidate palmprint recognition model by using the palmprint data and a data label corresponding to the palmprint data to obtain a target palmprint recognition model comprises:

inputting the palmprint data into the candidate palmprint recognition model, and determining a loss value corresponding to the palmprint data based on output data of the candidate palmprint recognition model and the data label corresponding to the palmprint data;

training the candidate palmprint recognition model by using the loss value; and

achieving a training target in response to the training on the candidate palmprint recognition model to obtain the target palmprint recognition model.

17 . An apparatus for generating a palmprint sample, comprising:

at least one processor and at least one memory, the at least one memory storing at least one instruction, at least one program, and a code set or an instruction set, and the at least one instruction, the at least one program, and the code set or the instruction set being loaded and executed by the at least one processor to implement:

generating positioning point data according to a palmprint principal line distribution rule, the positioning point data comprising first data corresponding to a first principal line positioning point and second data corresponding to a second principal line positioning point;

generating adjusting point data according to a palmprint principal line curving rule, a principal line adjusting point corresponding to the adjusting point data being configured to control a curving parameter of a principal line formed by the first principal line positioning point and the second principal line positioning point;

generating a palmprint principal line based on the first data, the second data and the adjusting point data; and

generating at least one palmprint sample comprising the palmprint principal line.

18 . The apparatus according to claim 17 , wherein the generating positioning point data according to a palmprint principal line distribution rule comprises:

determining a first region and a second region corresponding to a trend of the palmprint principal line according to the palmprint principal line distribution rule;

determining the first data corresponding to the first principal line positioning point in the first region; and

determining the second data corresponding to the second principal line positioning point in the second region.

19 . The apparatus according to claim 18 , wherein the determining a first region and a second region corresponding to a trend of the palmprint principal line according to the palmprint principal line distribution rule comprises:

determining a palmprint generation region according to the palmprint principal line distribution rule, the palmprint generation region indicating a distribution range of the palmprint principal line;

determining a first vertex and a second vertex in a diagonal relationship in the palmprint generation region;

determining the first region in the palmprint generation region by using the first vertex as a center and a first preset length as a radius; and

determining the second region in the palmprint generation region by using the second vertex as a center and a second preset length as a radius.

20 . A non-transitory computer-readable storage medium, storing at least one instruction, at least one program, and a code set or an instruction set, and the at least one instruction, the at least one program, and the code set or the instruction set being loaded and executed by at least one processor to implement:

generating positioning point data according to a palmprint principal line distribution rule, the positioning point data comprising first data corresponding to a first principal line positioning point and second data corresponding to a second principal line positioning point;

generating adjusting point data according to a palmprint principal line curving rule, a principal line adjusting point corresponding to the adjusting point data being configured to control a curving parameter of a principal line formed by the first principal line positioning point and the second principal line positioning point;

generating a palmprint principal line based on the first data, the second data and the adjusting point data; and

generating at least one palmprint sample comprising the palmprint principal line.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 17, 2023
From: ZHANG, YINGYI; ZHAO, KAI; SHEN, LEI; ZHANG, RUIXIN; ZHOU, CHUHAN; WANG, TAO; DING, SHOUHONG
To: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
Reel/Frame 064632/0596 →
Priority Claims (1)
CN 202210189742.8 · Feb 28, 2022 · national
Continuity (2)
Continuation PCTCN2022133470 · Nov 22, 2022
Related Publication 20230394869A1 · Dec 7, 2023
References Cited (40)
US 20160104030A1 · Matsunami · 2016 [cited by applicant]
CN 105512530A · 2016 [cited by applicant]
CN 108427923A · 2018 [cited by applicant]
CN 108615002A · 2018 [cited by applicant]
CN 109472213A · 2019 [cited by applicant]
CN 110163062A · 2019 [cited by applicant]
CN 110826452A · 2020 [cited by applicant]
CN 111950492A · 2020 [cited by examiner]
CN 112149521A · 2020 [cited by applicant]
WO 2019034589A1 · 2019 [cited by applicant]
The European Patent Office (EPO) The Extended European Search Report for Application No. 22928308.0 Feb. 27, 2025 12 Pages. [cited by applicant]
Thomas Cook. “the development of automated Palmprint identification using major flexion creases.” (2012). [cited by applicant]
Vani Perumal et al., “Pattern Recognition of Palm Print and Footprint Images for their Effective Storage Management—A Nonlinear Approach.” (2017). [cited by applicant]
Kai Zhao et al. “Geometric Synthesis: A Free lunch for Large-scale Palmprint Recognition Model Pretraining.” arXiv preprint arXiv:2203.05703 (2022). [cited by applicant]
Jianjiang Feng et al., “A preliminary study of handprint synthesis.” Chinese Conference on Biometric Recognition. Berlin, Heidelberg: Springer Berlin Heidelberg, 2011. [cited by applicant]
Zhuoshi Wei et al. “Palmprint image synthesis: A preliminary study.” 2008 15th IEEE International Conference on Image Processing. IEEE, 2008. [cited by applicant]
The World Intellectual Property Organization (WIPO) International Search Report for PCT/CN2022/133470 Jan. 18, 2023 7 Pages (including translation). [cited by applicant]
The State Intellectual Property Office of the People's Republic of China (SIPO) Office Action 1 for 202210189742.8, Apr. 7, 2023 8 Pages (including translation). [cited by applicant]
Nuoya Xu, Research on palmprint recognition in complex scenes, China Excellent Master's Thesis Full Text Database Information Technology Series, vol. 2021, Jul. 15, 2021. [cited by applicant]
Zhuoshi Wei et al., Palmprint image synthesis: A preliminary study, 2018 15th IEEE International Conference on Image Processing, Dec. 12, 2008. pp. 285-288. [cited by applicant]
Sheng Chen et al. “Mobilefacenets: Efficient cnns for accurate real-time face verification on mobile devices.” Biometric Recognition: 13th Chinese Conference, CCBR 2018, Urumqi, China, Aug. 11-12, 2018, Proceedings 13. … [cited by applicant]
Jia Deng et al. “Imagenet: A large-scale hierarchical image database.” 2009 IEEE conference on computer vision and pattern recognition. leee, 2009. [cited by applicant]
Jiankang Deng et al. “Arcface: Additive angular margin loss for deep face recognition.” Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019. [cited by applicant]
Miguel A.Ferrer et al., “BiSpectral contactless hand based biometric system.” CONATEL 2011. IEEE, 2011. [cited by applicant]
Angelo Genovese et al. “PalmNet: Gabor-PCA convolutional networks for touchless palmprint recognition.” IEEE Transactions on Information Forensics and Security 14.12 (2019): 3160-3174. [cited by applicant]
Ying Hao et al. “Multispectral palm image fusion for accurate contact-free palmprint recognition.” 2008 15th IEEE International Conference on Image Processing. IEEE, 2008. [cited by applicant]
Ahmad Hassanat et al. “New mobile phone and webcam hand images databases for personal authentication and identification.” Procedia Manufacturing 3 (2015): 4060-4067. [cited by applicant]
Kaiming He et al. “Deep residual learning for image recognition.” Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. [cited by applicant]
Vivek Kanhangad et al., “Contactless and pose invariant biometric identification using hand surface.” IEEE transactions on image processing 20.5 (2010): 1415-1424. [cited by applicant]
Ajay Kumar. “Incorporating cohort information for reliable palmprint authentication.” 2008 Sixth Indian conference on computer vision, graphics & image processing. IEEE, 2008. [cited by applicant]
Wojciech Michal Matkowski et al., “Palmprint recognition in uncontrolled and uncooperative environment.” IEEE Transactions on Information Forensics and Security 15 (2019): 1601-1615. [cited by applicant]
Huikai Shao et al., “Few-shot palmprint recognition via graph neural networks.” Electronics Letters 55.16 (2019): 890-892. [cited by applicant]
Huikai Shao et al., “Effective deep ensemble hashing for open-set palmprint recognition.” Journal of Electronic Imaging 29.1 (2020): 013018-013018. [cited by applicant]
Huikai Shao et al., “Deep distillation hashing for unconstrained palmprint recognition.” IEEE Transactions on Instrumentation and Measurement 70 (2021): 1-13. [cited by applicant]
Zhenan Sun et al. “Ordinal palmprint represention for personal identification [represention read representation].” 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR'05). vol. 1. IEEE… [cited by applicant]
David Zhang et al. “An online system of multispectral palmprint verification.” IEEE transactions on instrumentation and measurement 59.2 (2009): 480-490. [cited by applicant]
Yingyi Zhang et al. “Towards palmprint verification on smartphones.” arXiv preprint arXiv:2003.13266 (2020). [cited by applicant]
Dexing Zhong et al., “Centralized large margin cosine loss for open-set deep palmprint recognition.” IEEE Transactions on Circuits and Systems for Video Technology 30.6 (2019): 1559-1568. [cited by applicant]
Creases Synthesis: A Free Lunch for Palmprint Recognition, Anonymous CVPR 2022 Submission, Paper ID 1469. Confidential Review Copy. [cited by applicant]
COEP palm print database. https://www.coep.org.in/resources/coeppalmprintdatabase. [cited by applicant]