计算机辅助设计与图形学学报

结合注意力机制和邻域增强机制的牙颌模型分割网络

PDF下载

刘志华 ¹^,²⁾ , 薛久涛 ¹^,²⁾ , 唐浩 ³⁾ , 廖与禾 ¹^,²⁾^,^*

计算机辅助设计与图形学学报 | 图像视频处理 2025,37(12): 2181-2190

收起

计算机辅助设计与图形学学报 |图像视频处理 2025 , 37 (12) : 2181 -2190

结合注意力机制和邻域增强机制的牙颌模型分割网络

全屏

刘志华¹^,²⁾, 薛久涛¹^,²⁾, 唐浩³⁾, 廖与禾¹^,²⁾^,^*

作者信息

¹⁾西安交通大学现代设计及转子轴承系统教育部重点实验室　西安　710049

²⁾西安交通大学陕西省机械产品质量保障与诊断重点实验室　西安　710049

³⁾重庆科技大学机械与动力工程学院　重庆　401331

通讯作者:

（yhliao@xjtu.edu.cn）

作者简介:

刘志华（2000—），男，硕士研究生，主要研究方向为计算机视觉

薛久涛（1995—），男，博士研究生，主要研究方向为机械系统故障诊断

唐浩（1983—），男，博士，正高级工程师，硕士生导师，主要研究方向为工业数据挖掘

廖与禾（1973—），男，博士，副教授，博士生导师，论文通信作者，主要研究方向为机械系统故障诊断.

Segmentation Network of Dental Mesh Model Based on Attention Mechanism and Neighbor Enhancement

Zhihua Liu¹^,²⁾, Jiutao Xue¹^,²⁾, Hao Tang³⁾, Yuhe Liao¹^,²⁾^,^*

Affiliations

¹⁾Key Laboratory of Education Ministry for Modern Design and Rotor-Bearing System, Xi'an Jiaotong University, Xi'an　710049

²⁾Shaanxi Key Laboratory of Mechanical Product Quality Assurance and Diagnostics, Xi'an Jiaotong University, Xi'an　710049

³⁾School of Mechanical and Power Engineering, Chongqing University of Science and Technology, Chongqing　401331

出版时间: 2025-12-20 doi: 10.3724/SP.J.1089.2024-00037

文章导航

摘要

收起

针对现有牙齿模型深度学习分割网络过度依赖坐标特征，忽略了几何特征与坐标特征的差异性和互补作用的问题，提出一种结合注意力机制和邻域增强机制的牙颌模型分割网络. 首先采用一个注意力增强模块，通过不同的编码方式对输入特征进行邻域增强，并结合几何特征与坐标特征的性质分别设计差值注意力机制和自注意力机制，充分提取网格的上下文特征；其次构建互补性池化函数，融合最大池化、注意力池化以及平均池化促进不同特征间的互补作用，提高分割网络的鲁棒性；最后嵌入自适应注意力模块，在平衡多尺度特征间的数值差异的同时获得综合特征来实现网格分割. 在200个真实患者的牙颌模型数据集上与其他网络进行对比的实验结果表明，所提网络针对牙颌模型的分割准确率达96.18%；该网络优于对比的深度学习网络，具有更优的牙颌模型分割性能.

关键词

三维牙颌模型分割 / 邻域增强 / 语义分割网络 / 注意力机制

Abstract

收起

To address the issue of existing deep learning-based segmentation methods for dental models, which overly rely on coordinate features and overlook the differences and complementary nature of geometric and coordinate features, a segmentation network based on attention mechanism and neighbor enhancement is proposed. First, an attention-enhanced module is employed to enhance the neighborhood of input features through different encoding schemes. It designs a differential attention mechanism and a self-attention mechanism to extract comprehensive contextual features of the mesh by considering the properties of geometric and coordinate features separately. Next, a complementary pooling function is constructed to fuse maximum pooling, attention pooling, and average pooling, which promotes the complementary interaction among different features and enhances the robustness of the segmentation network. Finally, an adaptive attention module is embedded to balance the numerical differences among multi-scale features for effective mesh segmentation. Compared with other methods in 200 real patients dental model dataset, the segmentation accuracy of dental model is 96.18%. The experimental results show that the proposed method is superior to the compared deep learning networks and has better performance in the segmentation of dental models.

Key words

three-dimensional dental model segmentation / neighbor enhancement / semantic segmentation network / attention mechanism

引用本文

刘志华, 薛久涛, 唐浩, 廖与禾. 结合注意力机制和邻域增强机制的牙颌模型分割网络. 计算机辅助设计与图形学学报, 2025 , 37 (12) : 2181 -2190 . DOI: 10.3724/SP.J.1089.2024-00037

Zhihua Liu, Jiutao Xue, Hao Tang, Yuhe Liao. Segmentation Network of Dental Mesh Model Based on Attention Mechanism and Neighbor Enhancement[J]. Journal of Computer-Aided Design & Computer Graphics, 2025 , 37 (12) : 2181 -2190 . DOI: 10.3724/SP.J.1089.2024-00037

基金

收起

国家重点研发计划(2019YFB1311903)
国家自然科学基金面上项目(51575424)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

Sharifovich O S, Axmadjanovna Y U, Amrulloevich G S. Comprehensive treatment and diagnosis of functional and metabolic disorders in children with dental deformities[J]. Journal of Pharmaceutical Negative Results, 2023, 14(3): 3708-3715

[2]

Ural Ç, Kaleli N. Direct digitalization devices in today's dental practice: intra oral scanners[J]. Journal of Experimental and Clinical Medicine, 2021, 38(S2): 136-142

[3]

Wei G D, Cui Z M, Liu Y M, et al. TANet: towards fully automatic tooth arrangement[C]//Proceedings of the 16th European Conference on Computer Vision. Heidelberg: Springer, 2020: 481-497

[4]

Tong W H, Yang X K, Pan M D, et al. Spectral mesh segmentation via ℓ₀ gradient minimization[J]. IEEE Transactions on Visualization and Computer Graphics, 2020, 26(4): 1807-1820

[5]

Kakehbaraei S, Arvanaghi R, Seyedarabi H, et al. 3D tooth segmentation in cone-beam computed tomography images using distance transform[J]. Biomedical Signal Processing and Control, 2023, 79: Article No.104122

[6]

Zhou Z S, Wan H L, Zhang H Y, et al. Segmentation of spontaneous intracerebral hemorrhage on CT with a region growing method based on watershed preprocessing[J]. Frontiers in Neurology, 2022, 13: Article No.865023

[7]

Jiang Xiaotong, Dai Ning, Zhang Changdong, et al. A semi-automatic precise tooth segmentation algorithm of dental model[J]. Journal of Computer-Aided Design & Computer Graphics, 2020, 32(5): 820-829 (in Chinese) (姜晓通, 戴宁, 张长东, 等. 牙齿半自动精确分割算法[J]. 计算机辅助设计与图形学学报, 2020, 32(5): 820-829)

[8]

Kondo T, Ong S H, Foong K W C. Tooth segmentation of dental study models using range images[J]. IEEE Transactions on Medical Imaging, 2004, 23(3): 350-362

[9]

Wongwaen N, Sinthanayothin C. Computerized algorithm for 3D teeth segmentation[C]//Proceedings of the International Conference on Electronics and Information Engineering. Los Alamitos: IEEE Computer Society Press, 2010: V1-277-V1-280

[10]

Jiang B X, Zhang S Z, Shi M Y, et al. Alternate level set evolutions with controlled switch for tooth segmentation[J]. IEEE Access, 2022, 10: 76563-76572

[11]

Xu X J, Liu C, Zheng Y Y. 3D tooth segmentation and labeling using deep convolutional neural networks[J]. IEEE Transactions on Visualization and Computer Graphics, 2019, 25(7): 2336-2348

[12]

Tian S K, Dai N, Zhang B, et al. Automatic classification and segmentation of teeth on 3D dental model using hierarchical deep learning networks[J]. IEEE Access, 2019, 7: 84817-84828

[13]

Sun D Y, Pei Y R, Song G Y, et al. Tooth segmentation and labeling from digital dental casts[C]//Proceedings of the 17th IEEE International Symposium on Biomedical Imaging. Los Alamitos: IEEE Computer Society Press, 2020: 669-673

[14]

Lian C F, Wang L, Wu T H, et al. Deep multi-scale mesh feature learning for automated labeling of raw dental surfaces from 3D intraoral scanners[J]. IEEE Transactions on Medical Imaging, 2020, 39(7): 2440-2450

[15]

Wu T H, Lian C F, Lee S, et al. Two-stage mesh deep learning for automated tooth segmentation and landmark localization on 3D intraoral scans[J]. IEEE Transactions on Medical Imaging, 2022, 41(11): 3158-3166

[16]

Hu H Q, Li Z X, Gao W C. MPCNet: improved MeshSegNet based on position encoding and channel attention[J]. IEEE Access, 2023, 11: 23326-23334

[17]

Lian C F, Wang L, Wu T H, et al. MeshSNet: deep multi-scale mesh feature learning for end-to-end tooth labeling on 3D dental surfaces[C]//Proceedings of the 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention. Heidelberg: Springer, 2019: 837-845

[18]

Remmel T K, Extending morphological pattern segmentation to 3D voxels[J]. Landscape Ecology, 2022, 37(2): 373-380

[19]

Qi C R, Yi L, Su H, et al. PointNet++: deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc, 2017: 5105-5114

[20]

Zhang R T, Zhang G Y, Yin J H, et al. Mesh-based DGCNN: semantic segmentation of textured 3-D urban scenes[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: Article No.4402812

[21]

Ren D Y, Wu Z Y, Li J W, et al. Point attention network for point cloud semantic segmentation[J]. Science China Information Sciences, 2022, 65(9): Article No.192104

[22]

Wang Y, Tang X T, Yue C K. Enhancing the local graph semantic feature for 3D point cloud classification and segmentation[J]. IEEE Access, 2022, 10: 74620-74628

[23]

Liu C, Zeng D D, Akbar A, et al. Context-aware network for semantic segmentation toward large-scale point clouds in urban environments[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: Article No.5703915

[24]

Zeng Z Y, Xu Y Y, Xie Z, et al. LEARD-Net: semantic segmentation for large-scale point cloud scene[J]. International Journal of Applied Earth Observation and Geoinformation, 2022, 112: Article No.102953

[25]

Yin F K, Huang Z L, Chen T, et al. DCNet: large-scale point cloud semantic segmentation with discriminative and efficient feature aggregation[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2023, 33(8): 4083-4095

[26]

Zhang J D, Li C P, Song Q, et al. Automatic 3D tooth segmentation using convolutional neural networks in harmonic parameter space[J]. Graphical Models, 2020, 109: Article No.101071

[27]

Zheng Y Y, Chen B J, Shen Y F, et al. TeethGNN: semantic 3D teeth segmentation with graph neural networks[J]. IEEE Transactions on Visualization and Computer Graphics, 2023, 29(7): 3158-3168

[28]

Ma T, Yang Y Z, Zhai J C, et al. A tooth segmentation method based on multiple geometric feature learning[J]. Healthcare, 2022, 10(10): Article No.2089

[29]

Cui Z M, Li C J, Chen N L, et al. TSegNet: an efficient and accurate tooth segmentation network on 3D dental model[J]. Medical Image Analysis, 2021, 69: Article No.101949

[30]

Zhao Y, Zhang L M, Yang C S, et al. 3D Dental model segmentation with graph attentional convolution network[J]. Pattern Recognition Letters, 2021, 152: 79-85

[31]

Lin Z J, He Z S, Wang X, et al. DBGANet: dual-branch geometric attention network for accurate 3D tooth segmentation[OL]. [2024-01-17]. https://ieeexplore.ieee.org/document/10313281

[32]

Ben-Hamadou A, Smaoui O, Chaabouni-Chouayakh H, et al. Teeth3DS: a benchmark for teeth segmentation and labeling from intra-oral 3D scans[OL]. [2024-01-17]. https://arxiv.org/abs/2210.06094

[33]

Wu T H, Lian C F, Piers C, et al. Machine (deep) learning for orthodontic CAD/CAM technologies[M]//Ko C C, Shen D G, Wang L. Machine Learning in Dentistry. Cham: Springer, 2021: 117-129

2025年第37卷第12期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.3724/SP.J.1089.2024-00037

接收时间：2024-01-17
首发时间：2026-10-10
出版时间：2025-12-20

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2024-01-17
修回日期：2024-05-15

基金

国家重点研发计划(2019YFB1311903)

国家自然科学基金面上项目(51575424)

作者信息

¹⁾西安交通大学现代设计及转子轴承系统教育部重点实验室　西安　710049

²⁾西安交通大学陕西省机械产品质量保障与诊断重点实验室　西安　710049

³⁾重庆科技大学机械与动力工程学院　重庆　401331

通讯作者:

（yhliao@xjtu.edu.cn）

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/jsjfzsj/CN/10.3724/SP.J.1089.2024-00037

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT