In large-scale equipment manufacturing, such as high-precision assembly of aircraft and ships, traditional laser-based measurement methods face notable constraints: measurement cycles exceed 6 h, station transfer operations introduce cumulative errors, and overall deployment is costly. Similarly, conventional fixed-view camera systems require dense device placement to achieve full coverage, leading to high expenses and complex maintenance. These challenges significantly limit production efficiency and capacity. This paper addresses measurement difficulties in dynamically occluded environments by proposing an intelligent perception system based on visual sensors. The system integrates dynamic calibration and progressive localization techniques to deliver an efficient, cost-effective, and high-precision measurement solution. The core objective is to establish a robust system capable of covering a large-scale measurement volume of 31 × 17 × 9.1 m, ensuring a positioning error of ≤2 mm, reducing response time to the order of seconds, and providing a reusable technical framework for industrial applications.
DaiRLiWTangX, et al. High-accuracy calibration for a multiview microscopic 3-D measurement system. IEEE/ASME Trans Mechatron2023; 29(4): 2475–2486
2.
ChenHLiuGWangZ. A stereovision-based efficient measurement approach for surface flatness of concrete members. Structures2024; 63: 106374.
3.
TaoLXiaRZhaoJ, et al. A high-accuracy circular hole measurement method based on multi-camera system. Measurement2023; 207: 112361.
4.
YuZShenGZhaoZ, et al. An improved method of concentric circle positioning in visual measurement. Opt Commun2023; 544: 129620.
5.
HaoFSuJShiJ, et al. Conic tangents based high precision extraction method of concentric circle centers and its application in camera parameters calibration. Sci Rep2021; 11(1): 20686.
6.
HeBZhaoZShaoW, et al. Identification and localization of circular coded target under non-uniform illumination. Measurement2025; 247: 116769.
7.
ChengXLiuZZengZ, et al. Precise camera calibration method based on eccentricity correction of circular target. Meas Sci Technol2025; 36(3): 035007.
8.
LemiaoYFuqiangZWanningZ. A novel camera calibration method based on circle projection model. Measurement2023; 222: 45–67.
9.
WangJWuHZhuZ, et al. Impact of speckle deformability on digital imaging correlation. IEEE Access2024; 12: 66466–66477.
10.
LiuZGuanBShangY, et al. LECalib: line-based event camera calibration. Measurement2024; 235: 114900.
11.
LiangSGuanBYuZ, et al. Camera calibration using a collimator system. In: European conference on computer vision, 2024, pp.374–390. Cham, Switzerland: Springer Nature.
12.
LiuWZhuJLiW, et al. High-precision stereo calibration for 3D measurement via speckle-enhanced circular control point matching. Appl Opt2025; 64(13): 3561–3570.
13.
LindenbergerPSarlinPEPollefeysM. LightGlue: local feature matching at light speed. In: Proceedings of the IEEE/CVF international conference on computer vision, Paris, France, 1–6 October 2023, pp.17627–17638. New York: IEEE.
14.
WangSKannalaJBarathD. DGC–GNN: leveraging geometry and color cues for visual descriptor-free 2D–3D matching. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Seattle, WA, USA, 16–22 June 2024, pp.20881–20891. New York: IEEE.
15.
SarlinPEUnagarALarssonM, et al. Back to the feature: learning robust camera localization from pixels to pose. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Nashville, TN, USA, 20–25 June 2021, pp.3247–3257. New York: IEEE.
16.
BrachmannECavallariTPrisacariuVA. Accelerated coordinate encoding: learning to relocalize in minutes using RGB and poses. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Vancouver, BC, Canada, 17–24 June 2023, pp.5044–5053. New York: IEEE.
17.
DongSWangSZhuangY, et al. Visual localization via few-shot scene region classification. In: 2022 International conference on 3D vision (3DV), Prague, Czech Republic, 12–16 September 2022, pp.393–402. New York: IEEE.
18.
ShavitYFerensRKellerY. Coarse-to-fine multi-scene pose regression with transformers. IEEE Trans Pattern Anal Mach Intell2023; 45(12): 14222–14233.
19.
YinYWangYWangH, et al. A Laplace-inspired distribution on SO(3) for probabilistic rotation estimation. arXiv preprint arXiv:2303.01743. 2023.
20.
ChenSBhalgatYLiX, et al. Neural refinement for absolute pose regression with feature synthesis. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Seattle, WA, USA, 16–22 June 2024, pp.20987–20996. New York: IEEE.
21.
LinJGuJWuB, et al. Learning neural volumetric pose features for camera localization. In: European conference on computer vision, Milan, Italy, 29 September–4 October 2024, pp.198–214. Cham: Springer Nature.
22.
MoreauAPiascoNTsishkouD, et al. Lens: localization enhanced by NeRF synthesis. In: Conference on robot learning, Cambridge, MA, January 2022, pp.1347–1356. PMLR.
23.
ArnoldEWynnJVicenteS, et al. Map-free visual relocalization: metric pose relative to a single image. In: European conference on computer vision, Tel Aviv, Israel, 23–27 October 2022, pp.690–708. Cham: Springer Nature.
24.
BaiYGengXMangalamK, et al. Sequential modeling enables scalable learning for large vision models. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Seattle, WA, USA, 16–22 June 2024, pp.22861–22872. New York: IEEE.
25.
KirillovAMintunERaviN, et al. Segment anything. In: Proceedings of the IEEE/CVF international conference on computer vision, Paris, France, 1–6 October 2023, pp.4015–4026. New York: IEEE.
26.
HeKChenXXieS, et al. Masked autoencoders are scalable vision learners. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, New Orleans, LA, USA, 18–24 June 2022, pp.16000–16009. New York: IEEE.
27.
DongSWangSLiuS, et al. Reloc3r: large-scale training of relative camera pose regression for generalizable, fast, and accurate visual localization. In: Proceedings of the computer vision and pattern recognition conference, Nashville, TN, USA, 10–17 June 2025, pp.16739–16752. New York: IEEE.
28.
WengYQuekSTYeohJKW. Robust vision-based sub-pixel level displacement measurement using a complementary strategy. Mech Syst Signal Process2025; 223: 111898.
29.
ZhangCLuZLiX, et al. A two-stage correction method for UAV movement-induced errors in non-target computer vision-based displacement measurement. Mech Syst Signal Process2025; 224: 112131.
30.
NiuYLiZLiJ, et al. Accelerometer-assisted computer vision data fusion framework for structural dynamic displacement reconstruction. Measurement2025; 242: 116021.
31.
NusantikaNRXiaoJHuX. Three-dimensional measurement of transmission line icing based on a rule-based stereo vision framework. Electronics2025; 14(21): 4184.
