三维内窥镜技术概述

Overview of 3D Endoscopic Technology

摘要

三维内窥镜（3D Endoscope）是依托立体视觉技术的无损检测设备，借助双镜头、结构光或激光扫描达成三维成像，弥补传统2D内窥镜没有深度信息的不足。它的核心技术涵盖立体匹配算法、实时点云重建和高精度深度测量，能量化缺陷尺寸，大量用于航空发动机、管道焊缝等复杂工业场景，以及微创手术等医疗领域。和2D内窥镜比，它在空间感知能力和数据可测量性上有优势，搭配AI算法可自动识别缺陷并生成三维模型。目前发展重点放在轻量化、无线化及多模态融合，不过存在成本高、数据处理复杂等难题。后续，三维内窥镜将朝智能化、便携化方向发展，成为工业4.0和精准医疗的核心工具。

Abstract

The 3D endoscope is a non-destructive inspection device that leverages stereoscopic vision technology to achieve three-dimensional imaging through dual-lens systems, structured light, or laser scanning, thereby compensating for the lack of depth information inherent in traditional 2D endoscopes. Its core technologies encompass stereoscopic matching algorithms, real-time point cloud reconstruction, and high-precision depth measurement; these capabilities enable the quantification of defect dimensions and find extensive applications in complex industrial scenarios such as aerospace engines and pipeline welds, as well as in medical fields like minimally invasive surgery. Compared with 2D endoscopes, the 3D endoscope offers advantages in spatial perception and data measurability; when integrated with AI algorithms, it can automatically identify defects and generate 3D models. Current research and development efforts are focused on lightweight design, wireless connectivity, and multimodal fusion; however, challenges such as high costs and complex data processing remain. In the future, 3D endoscopes will evolve toward greater intelligence and portability, becoming core tools for Industry 4.0 and precision medicine.

关键词

内窥镜 / 三维模型 / 三维内窥镜

Key words

endoscope / 3D model / 3D endoscope

参考文献

[1] 刘腾云. 胶囊内窥镜图像特征点提取与匹配方法研究[D]. 江南大学, 2025.

[2] 袁晓涵. 窑炉监测工业内窥镜设计[D]. 长春理工大学, 2025.

[3] Yoshimoto K, Watabe K, Fujinaga T, et al. Evaluation of endoscopic entire 3D image acquisition of the digestive tract using a stereo endoscope[C]// SPIE, 2017: 10.1117/12.2250774.

[4] Lee S J, Jung S G, Won H Y. Light field 3D endoscope based on electro-wetting lens array[C]// SPIE, 2017: 10.1117/12.2251660.

[5] Le D N H, Nguyen H, Wang Z, et al. 3D endoscopic imaging using structured illumination technique (Conference Presentation)[C]// SPIE, 2017: 10.1117/12.2250242.