基于三维耳廓模板修复光学扫描缺损模型的方法研究

doi:10.11996/JG.j.2095-302X.2025040889

图学学报 ›› 2025, Vol. 46 ›› Issue (4): 889-898.DOI: 10.11996/JG.j.2095-302X.2025040889

基于三维耳廓模板修复光学扫描缺损模型的方法研究

林智源¹(), 王业维², 余光正¹^,²(), 李哲林¹, 张鑫³

1.华南理工大学设计学院，广东广州 510006
2.华南理工大学物理与光电学院，广东广州 510641
3.华南理工大学电子与信息学院，广东广州 510641

收稿日期:2025-01-13 修回日期:2025-04-15 出版日期:2025-08-30 发布日期:2025-08-11
通讯作者:余光正(1978-)，男，教授，博士。主要研究方向为声学穿戴产品设计、双耳听觉等。E-mail：scgzyu@scut.edu.cn
第一作者:林智源(2000-)，男，硕士研究生。主要研究方向为声学穿戴产品设计、人机工效学设计。E-mail：202221055927@mail.scut.edu.cn
基金资助:
国家自然科学基金(12474465);国家重点研发计划(2022YFF0607000);广东省自然科学基金面上项目(2024A1515011446)

Research on the method of repairing optical scanning incomplete model based on three-dimensional auricle template

LIN Zhiyuan¹(), WANG Yewei², YU Guangzheng¹^,²(), LI Zhelin¹, ZHANG Xin³

1. School of Design, South China University of Technology, Guangzhou Guangdong 510006, China
2. School of Physics and Optoelectronics, South China University of Technology, Guangzhou Guangdong 510641, China
3. School of Electronic and Information Engineering, South China University of Technology, Guangzhou Guangdong 510641, China

Received:2025-01-13 Revised:2025-04-15 Published:2025-08-30 Online:2025-08-11
First author：LIN Zhiyuan (2000-), master student. His main research interests cover acoustic wearable product design, ergonomic design. E-mail：202221055927@mail.scut.edu.cn
Supported by:
National Natural Science Foundation of China(12474465);National Key Research and Development Program of China(2022YFF0607000);Natural Science Foundation of Guangdong Province(2024A1515011446)

摘要/Abstract

摘要： 人体耳廓的三维模型在人体工程学和数值模拟等领域有着重要应用，可通过光学扫描成像等方法快速建模，但耳甲腔等局部凹陷结构无法将反射光线反馈给扫描设备，从而形成扫描盲区。传统方法是通过逆向工程软件依据盲区周围三角网格的曲率信息进行修补，但精度较低且效率不高。为提升耳廓模型精度，将耳印材料注入耳甲腔，然后扫描耳印模型以获得精确的耳甲腔凹陷结构，并将其与耳廓扫描模型的重合部位进行手动对齐和全局对齐，从而拼接得到较为精确且完整模型，该方法获得的耳廓模型精确但整个过程相对复杂耗时。针对这一问题，提出一种基于三维耳廓模板修复光学扫描缺损模型的方法。首先基于精确扫描和拼接建立的35名成年受试者的耳廓模型构建统计形状模型库，根据其相关性特征生成三维耳廓模板；然后采用改进后的MeshMonk程序，通过刚性配准和非刚性配准至5位新受试者的缺损模型，从而生成完整模型；最后采用偏差距离均方根值，对比分析5位受试者的修补完整模型、生成完整模型和精确完整模型之间的图形偏差。结果表明，生成完整模型与精确完整模型的均方根均值为(0.37±0.01) mm，在可接受阈值0.50 mm范围内。而在点云距离大于0.50 mm之外，生成完整模型与精确完整模型的均方根误差均值为(0.93±0.12) mm，明显小于修补完整模型与精确完整模型的均方根误差均值(2.87±0.49) mm (提高约68%)。可见，该方法相较于逆向工程软件修补法具有明显的精度优势，且方法简洁、高效，可用于完整耳廓甚至头部的三维扫描建模。

关键词: 光学扫描, 缺损模型, 三维耳廓模板, 模型配准, 精度对比

Abstract:

Three-dimensional models of the human auricle are essential for ergonomics and numerical simulation. Optical scanning imaging enables fast modeling, but local recessed structures, such as the cavum concha, cannot reflect light back to the scanning device, resulting in scanning blind spots. Traditional methods use reverse-engineering software to patch these areas based on the curvature of surrounding triangular meshes; however, the accuracy is relatively low and the efficiency is not high. To enhance the model accuracy, ear impression material can be injected into the cavum concha, and the resulting ear impression model is then scanned to obtain the precise recessed structure. This model is manually aligned and globally registered with the auricle scanning model at their overlapping regions to generate a relatively accurate and complete model. However, the overall process is complex and time-consuming. To address this issue, a method based on a three-dimensional auricle template to repair optical scanning incomplete models was proposed. First, the method constructed a statistical shape model library based on accurate complete models from 35 adult subjects and generated templates based on their correlated features. Then, an improved MeshMonk program was employed to rigidly and non-rigidly register the templates to the incomplete models of 5 new subjects to generate complete models. Finally, the root mean square (RMS) of deviation distances was used to compare and analyze the geometric differences among the repaired complete model, the generated complete model, and the accurate complete model for five subjects. The results indicated that the RMS error between the generated complete model and the accurate complete model was (0.37±0.01) mm (within the acceptable threshold of 0.50 mm). For point clouds with distances exceeding 0.50 mm, the RMS error between the generated complete model and the accurate complete model was (0.93±0.12) mm, smaller than the RMS error between the repaired complete model and the accurate complete model (2.87±0.49) mm. This demonstrated that the proposed method improved accuracy by approximately 68% compared to the reverse engineering software repair method when the point cloud distance exceeded 0.50 mm. Moreover, the method was streamlined and efficient, suggesting applicability for 3D scanning and modeling of the entire auricle or even the head.

Key words: optical scanning, incomplete model, 3D auricle template, model registration, accuracy comparison

中图分类号:

林智源, 王业维, 余光正, 李哲林, 张鑫. 基于三维耳廓模板修复光学扫描缺损模型的方法研究[J]. 图学学报, 2025, 46(4): 889-898.

LIN Zhiyuan, WANG Yewei, YU Guangzheng, LI Zhelin, ZHANG Xin. Research on the method of repairing optical scanning incomplete model based on three-dimensional auricle template[J]. Journal of Graphics, 2025, 46(4): 889-898.

图/表 13

表1 三维扫描仪及相关参数

Table 1 3D scanner and related parameters

参数	扫描仪
参数	手持式	桌面级
商品名称	Reeyee Pro 2X	Reeyee
光源	三色LED	白光LED
扫描范围	360°	360°
精度/mm	0.1	0.1
分辨率/mm	0.50，1.00，1.50	0.17~0.20
配套软件	Reeyee Pro_v3.4.0.1	Reeyee_v2.7.1.0
输出格式	OBJ，STL，PLY	OBJ，STL，PLY

图1 精确完整模型MaC采集流程

Fig. 1 Acquisition process of MaC

图2 创建三维耳廓模板的部分流程((a) 对齐至全局坐标系；(b) 耳廓提取结果)

Fig. 2 Partial process of creating a three-dimensional auricle template ((a) Align to the global coordinate system; (b) Auricle extraction results)

图3 三维耳廓模板Mtemp

Fig. 3 Three-dimensional auricle template

图4 MeshMonk表面配准算法示意图

Fig. 4 Schematic diagram of the MeshMonk surface registration algorithm

图5 用于耳廓注册的地标点

Fig. 5 Landmark points for auricle registration ((a) MiC; (b) Mtemp)

图6 三维耳廓模板法修复耳廓缺损模型过程

Fig. 6 The process of repairing the auricular incomplete model by three-dimensional auricular template method

图7 MiC通过3种方法处理后的曲面形态((a) 缺损模型；(b) 软件修补法；(c) 模板配准法；(d) 耳印补偿法)

Fig. 7 Surface morphologies of MiC after treatment by three methods ((a) Incomplete model; (b) Software patching method; (c) Template registration method; (d) Ear print compensation method)

图8 MgC, MrC与MaC的距离色阶图

Fig. 8 Distance color scale diagram of MgC, MrC and MaC ((a) MgC-MgC; (b) MrC-MgC)

表2 生成完整模型与修补完整模型的均方根值比较

Table 2 Comparison of root mean square values between MgC and MrC

受试者	生成完整模型/mm	修补完整模型/mm
1	0.36	0.54
2	0.38	0.54
3	0.38	0.56
4	0.36	0.48
5	0.37	0.36
均值±标准差	0.37±0.01^a	0.50±0.08^a

图9 MgC和MaC与MrC和MaC的点云距离频率直方图((a) 距离≤-0.5 mm；(b) 距离≥0.5 mm)

Fig. 9 Point cloud distance frequency histograms of MgC and MaC, MrC and MaC ((a) ≤-0.5 mm; (b) ≥0.5 mm)

表3 生成完整模型与修补完整模型的均方根误差值比较

Table 3 Comparison of root mean square values between MgC and MrC

受试者	生成完整模型/mm	修补完整模型/mm
1	0.76	2.88
2	1.02	2.23
3	0.84	3.41
4	1.03	3.28
5	1.01	2.54
均值±标准差	0.93±0.12^a	2.87±0.49^a

图10 均方根误差值箱型图

Fig. 10 Box plot of root mean square error values

参考文献 42

[1]	RODRÍGUEZ-ARIAS J P, GUTIÉRREZ VENTURINI A, PAMPÍN MARTÍNEZ M M, et al. Microtia ear reconstruction with patient-specific 3D models—a segmentation protocol[J]. Journal of Clinical Medicine, 2022, 11(13): 3591.
[2]	NIGHTINGALE R C, ROSS M T, CRUZ R L J, et al. Frugal 3D scanning using smartphones provides an accessible framework for capturing the external ear[J]. Journal of Plastic, Reconstructive & Aesthetic Surgery, 2021, 74(11): 3066-3072.
[3]	FAN X M, NIU X M, WANG Y B, et al. Comparison of three-dimensional and two-dimensional templates on auricle reconstruction in patients with unilateral microtia[J]. American Journal of Translational Research, 2019, 11(6): 3771-3778. PMID
[4]	FAN H, YU S H, WANG M C, et al. Analysis of the external acoustic meatus for ergonomic design: part I-measurement of the external acoustic meatus using casting, scanning and rapid estimation approaches[J]. Ergonomics, 2021, 64(5): 640-656.
[5]	BAN K, JUNG E S. Ear shape categorization for ergonomic product design[J]. International Journal of Industrial Ergonomics, 2020, 80: 102962.
[6]	LEE W, YANG X P, JUNG H, et al. Anthropometric analysis of 3D ear scans of Koreans and Caucasians for ear product design[J]. Ergonomics, 2018, 61(11): 1480-1495. DOI PMID
[7]	ZHU Z H, JI X M, GAO Z, et al. A morphometric study of auricular concha in the population of young Chinese adults[J]. International Journal of Morphology, 2017, 35(4): 1451-1458.
[8]	PAVLIDIS G, KOUTSOUDIS A, ARNAOUTOGLOU F, et al. Methods for 3D digitization of cultural heritage[J]. Journal of Cultural Heritage, 2007, 8(1): 93-98.
[9]	孙启翔. 三维点云数据的孔洞修补技术研究[D]. 长春: 长春工业大学, 2024.
	SUN Q X. Research on hole repairing technology for 3D point cloud data[D]. Changchun: Changchun University of Technology, 2024 (in Chinese).
[10]	CHO S D, JANG J H, KIM H, et al. Earmold foreign bodies in the middle ear necessitating surgical removal: why otology specialists should screen candidates for hearing aids[J]. Clinical and Experimental Otorhinolaryngology, 2021, 14(2): 235-239.
[11]	CHEN C K, HSIEH L C, CHIANG Y C, et al. Feasibility of high‐resolution computed tomography imaging for obtaining ear impressions for hearing aid fitting[J]. Otolaryngology- Head and Neck Surgery, 2019, 161(4): 666-671.
[12]	NOLAN M, COMBE E C. Silicone materials for ear impressions[J]. Scandinavian Audiology, 1985, 14(1): 35-39. PMID
[13]	范然, 金小刚. 大规模点云选择及精简[J]. 图学学报, 2013, 34(3): 12-19.
	FAN R, JIN X G. Selection and reduction algorithms for large point clouds[J]. Journal of Graphics, 2013, 34(3): 12-19 (in Chinese).
[14]	袁天然. 三角网格模型光顺、简化和缝补技术的研究及应用[D]. 南京: 南京航空航天大学, 2007.
	YUAN T R. Research & application of triangular mesh model’s hole filling, decimation and stitching algorithm[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2007 (in Chinese).
[15]	DEMUYNCK M, DELNAVAZ A, VOIX J. Morphological analysis of the human earcanal deformations during face- related activities[J]. Applied Ergonomics, 2024, 116: 104195.
[16]	FAN H, YU S H, WANG M C, et al. Analysis of the external acoustic meatus for ergonomic design: part II-anthropometric variations of the external acoustic meatus by sex, age and side in Chinese population[J]. Ergonomics, 2021, 64(5): 657-670.
[17]	LEE W, JUNG H, BOK I, et al. Measurement and application of 3D ear images for earphone design[J]. Proceedings of The Human Factors and Ergonomics Society Annual Meeting, 2016, 60(1): 1053-1057.
[18]	VERDAM F, TANGE R, THOMEER H. Impression material in the external and middle ear: an overview of the literature and a stepwise approach for removal[J]. The Journal of International Advanced Otology, 2016, 12(3): 345-352.
[19]	LEE H M, YI K I, JUNG J H, et al. Hearing aid silicone impression material as a foreign body in the middle ear[J]. American Journal of Otolaryngology, 2017, 38(1): 108-111.
[20]	YOSHIOKA S, INUZUKA E, KATO H, et al. Surgical removal of hearing aid ear mold impression material from the middle ear: a report of two cases[J]. Fujita Medical Journal, 2017, 3(3): 72-75.
[21]	VAN DEN BOER C, VAN SPRONSEN E, HOLLAND C T Q, et al. Clinical approach after complicated ear mold fitting: a case series of six patients and evaluation of literature[J]. Annals of Otology, Rhinology & Laryngology, 2019, 128(12): 1141-1146.
[22]	NOLAN M, COMBE E C. In vitro considerations in the production of dimensionally accurate earmoulds I.: the ear impression[J]. Scandinavian Audiology, 1989, 18(1): 35-41.
[23]	AMBERG B, ROMDHANI S, VETTER T. Optimal step nonrigid ICP algorithms for surface registration[C]// 2007 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2007: 1-8.
[24]	杨荣, 冯有前, 袁修久. 利用现有模型修复不完整三维模型[J]. 计算机辅助设计与图形学学报, 2015, 27(1): 98-105.
	YANG R, FENG Y Q, YUAN X J. Restoration of fragmentary 3D models using existing models[J]. Journal of Computer-Aided Design & Computer Graphics, 2015, 27(1): 98-105 (in Chinese).
[25]	吴云燕, 陈正鸣, 何坤金. 网格分割及形态参数引导的股骨模型修复[J]. 计算机辅助设计与图形学学报, 2018, 30(10): 1899-1909.
	WU Y Y, CHEN Z M, HE K J, et al. Repair of defected femur models based on mesh segmentation and morphological parameters[J]. Journal of Computer-Aided Design & Computer Graphics, 2018, 30(10): 1899-1909 (in Chinese).
[26]	ISO. ISO 15535:2023 General requirements for establishing anthropometric databases[S]. Geneva, Switzerland: ISO, 2023.
[27]	WANG M C, FAN H, YU S H, et al. Analysis of the auricles and auricular shape types for ear-related wearables: a study of mainland Chinese sample aged 15-79[J]. Work, 2022, 73(1): 335-352. DOI PMID
[28]	吴怡, 董可欣, 杨峥, 等. 不同年龄耳廓来源软骨细胞的生物学特性研究[J]. 中华整形外科杂志, 2019, 35(4): 331-340.
	WU Y, DONG K X, YANG Z, et al. Biological characteristics of chondrocytes derived from auricles of different ages[J]. Chinese Journal of Plastic Surgery, 2019, 35(4): 331-340 (in Chinese).
[29]	ZHAO S C, LI D G, LIU Z Z, et al. Anthropometric growth study of the ear in a Chinese population[J]. Journal of Plastic, Reconstructive & Aesthetic Surgery, 2018, 71(4): 518-523.
[30]	GUO Z Y, LU Y G, ZHOU H L, et al. Anthropometric-based clustering of pinnae and its application in personalizing HRTFs[J]. International Journal of Industrial Ergonomics, 2021, 81: 103076.
[31]	FU F, LUXIMON A, LUXIMON Y. 3D human ear modelling with parameterization technique and variation analysis[J]. Ergonomics, 2024, 67(5): 638-649.
[32]	WHITE J D, ORTEGA-CASTRILLÓN A, MATTHEWS H, et al. MeshMonk: open-source large-scale intensive 3D phenotyping[J]. Scientific Reports, 2019, 9(1): 6085. DOI PMID
[33]	BERENDS B, BIELEVELT F, SCHREURS R, et al. Fully automated landmarking and facial segmentation on 3D photographs[J]. Scientific Reports, 2024, 14(1): 6463. DOI PMID
[34]	BAI Y K, TRAN H, DAMELIN S B, et al. Partial transport for point-cloud registration[J]. Sampling Theory, Signal Processing, and Data Analysis, 2025, 23(1): 4.
[35]	LOW K L. Linear least-squares optimization for point-to-plane icp surface registration[R]. Chapel Hill: University of North Carolina, 2004.
[36]	LATHAM J, LUDLOW M, MENNITO A, et al. Effect of scan pattern on complete-arch scans with 4 digital scanners[J]. The Journal of Prosthetic Dentistry, 2020, 123(1): 85-95.
[37]	吴亦奇, 何嘉乐, 张甜甜, 等. 基于多重特征提取和点对应关系的三维点云非刚配准[J]. 图学学报, 2025, 46(1): 150-158. DOI
	WU Y Q, HE J L, ZHANG T T, et al. Unsupervised 3D point cloud non-rigid registration based on multi-feature extraction and point correspondence[J]. Journal of Graphics, 2025, 46(1): 150-158 (in Chinese). DOI
[38]	GIBELLI D, PALAMENGHI A, POPPA P, et al. Improving 3D-3D facial registration methods: potential role of three-dimensional models in personal identification of the living[J]. International Journal of Legal Medicine, 2021, 135(6): 2501-2507. DOI PMID
[39]	陈国军, 曹岳, 杨静, 等. 基于形变模型的多角度三维人脸实时重建[J]. 图学学报, 2019, 40(4): 659-664. DOI
	CHEN G J, CAO Y, YANG J, et al. Real-time reconstruction of multi-angle 3D human faces based on morphable model[J]. Journal of Graphics, 2019, 40(4): 659-664 (in Chinese).
[40]	REVILLA-LEÓN M, GÓMEZ-POLO M, BARMAK A B, et al. Influence of occlusal collision corrections completed by two intraoral scanners or a dental design program on the accuracy of the maxillomandibular relationship[J]. The Journal of Prosthetic Dentistry, 2024, 132(1): 191-203.
[41]	SHARMA N, CAO S S, MSALLEM B, et al. Effects of steam sterilization on 3D printed biocompatible resin materials for surgical guides—an accuracy assessment study[J]. Journal of Clinical Medicine, 2020, 9(5): 1506.
[42]	FAROOK T H, RASHID F, JAMAYET N B, et al. A virtual analysis of the precision and accuracy of 3-dimensional ear casts generated from smartphone camera images[J]. The Journal of Prosthetic Dentistry, 2022, 128(4): 830-836.

基于三维耳廓模板修复光学扫描缺损模型的方法研究

Research on the method of repairing optical scanning incomplete model based on three-dimensional auricle template

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 42

相关文章 1

编辑推荐

Metrics

本文评价