Dental Identification Based on Single Tooth Feature Concatenation

Forensic dentistry research indicates that teeth, as one of the hardest tissues in the human body, can be well-preserved even in disasters and crime scenes, providing a reliable means of human identification. Dental identification based on 3D tooth point clouds has made some progress recently. But t...

Full description

Saved in:
Bibliographic Details
Main Authors: Li Yuan, Luoji Zhu, Kang-Hyun Jo, Yanfeng Li
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11071270/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849714635149672448
author Li Yuan
Luoji Zhu
Kang-Hyun Jo
Yanfeng Li
author_facet Li Yuan
Luoji Zhu
Kang-Hyun Jo
Yanfeng Li
author_sort Li Yuan
collection DOAJ
description Forensic dentistry research indicates that teeth, as one of the hardest tissues in the human body, can be well-preserved even in disasters and crime scenes, providing a reliable means of human identification. Dental identification based on 3D tooth point clouds has made some progress recently. But the whole-dentition-based identification mode may fail in complex forensic application scenarios such as tooth loss, changes in arrangement or scattering. Therefore, we propose a dental identification deep neural network based on single tooth feature concatenation in this paper, offering more flexibility and accuracy as it can adapt to various tooth arrangements and different tooth numbers. To address the challenges of difficult samples with fewer teeth, a progressive random tooth dropout training strategy is devised to aid learning. For inadequate discrimination of individual tooth features, PointResNet is introduced as a feature extraction module. To handle significant differences in samples with varying tooth numbers and arrangements from the same individual, the additive angular margin loss is adopted. To verify the effectiveness of our method in various dental conditions, we apply random spatial transformations to the teeth of the same individual and simulate different cases of missing teeth on our self-constructed dataset. The results from comparative and ablation experiments confirmed the effectiveness of this approach.
format Article
id doaj-art-3976f05af64144088d26f3f13d42e1a6
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-3976f05af64144088d26f3f13d42e1a62025-08-20T03:13:39ZengIEEEIEEE Access2169-35362025-01-011312279212280310.1109/ACCESS.2025.358554511071270Dental Identification Based on Single Tooth Feature ConcatenationLi Yuan0https://orcid.org/0009-0007-6172-6322Luoji Zhu1https://orcid.org/0000-0001-9836-7115Kang-Hyun Jo2https://orcid.org/0000-0002-4937-7082Yanfeng Li3School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, ChinaSchool of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, ChinaDepartment of Electrical, Electronic and Computer Engineering, University of Ulsan, Ulsan, South KoreaDepartment of Stomatology, Chinese PLA General Hospital Fourth Medical Center, Beijing, ChinaForensic dentistry research indicates that teeth, as one of the hardest tissues in the human body, can be well-preserved even in disasters and crime scenes, providing a reliable means of human identification. Dental identification based on 3D tooth point clouds has made some progress recently. But the whole-dentition-based identification mode may fail in complex forensic application scenarios such as tooth loss, changes in arrangement or scattering. Therefore, we propose a dental identification deep neural network based on single tooth feature concatenation in this paper, offering more flexibility and accuracy as it can adapt to various tooth arrangements and different tooth numbers. To address the challenges of difficult samples with fewer teeth, a progressive random tooth dropout training strategy is devised to aid learning. For inadequate discrimination of individual tooth features, PointResNet is introduced as a feature extraction module. To handle significant differences in samples with varying tooth numbers and arrangements from the same individual, the additive angular margin loss is adopted. To verify the effectiveness of our method in various dental conditions, we apply random spatial transformations to the teeth of the same individual and simulate different cases of missing teeth on our self-constructed dataset. The results from comparative and ablation experiments confirmed the effectiveness of this approach.https://ieeexplore.ieee.org/document/11071270/Dental identification3D tooth point cloudsingle tooth feature concatenationPointResNet
spellingShingle Li Yuan
Luoji Zhu
Kang-Hyun Jo
Yanfeng Li
Dental Identification Based on Single Tooth Feature Concatenation
IEEE Access
Dental identification
3D tooth point cloud
single tooth feature concatenation
PointResNet
title Dental Identification Based on Single Tooth Feature Concatenation
title_full Dental Identification Based on Single Tooth Feature Concatenation
title_fullStr Dental Identification Based on Single Tooth Feature Concatenation
title_full_unstemmed Dental Identification Based on Single Tooth Feature Concatenation
title_short Dental Identification Based on Single Tooth Feature Concatenation
title_sort dental identification based on single tooth feature concatenation
topic Dental identification
3D tooth point cloud
single tooth feature concatenation
PointResNet
url https://ieeexplore.ieee.org/document/11071270/
work_keys_str_mv AT liyuan dentalidentificationbasedonsingletoothfeatureconcatenation
AT luojizhu dentalidentificationbasedonsingletoothfeatureconcatenation
AT kanghyunjo dentalidentificationbasedonsingletoothfeatureconcatenation
AT yanfengli dentalidentificationbasedonsingletoothfeatureconcatenation