.png){kind=icon}
.png){kind=icon}
Return
An enhanced visual transfer calibration method for global measurement of irregular-shaped workpieces
.png){kind=icon}
PRE
AI
G
Z
W
H
L
DOI：10.1088/1361-6501/ae6680.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Achieving both a large measurement range and high measurement accuracy remains a major challenge of optical measurement of large irregular-shaped workpieces. To address this issue, this paper proposes an enhanced visual transfer calibration method for global dimensional measurement. Different from the traditional approaches that rely on full field-of-view (FOV) coverage, the proposed method adopts a 'local measurement-spatial association' method. High-resolution image data of the measured region are first captured within a small optical FOV, allowing full exploitation of the local precision characteristics of computer vision systems and thereby achieving local resolution enhancement. A high-precision global spatial coordinate transformation model is then constructed to map the local measurement results into a unified global reference frame, effectively expanding the measurable range and completing full-system calibration. Results show that the system achieves a mean error below 0.02 mm, a standard deviation below 0.07 mm, further bias and linearity analysis verify that the measurement results comply with the required task accuracy of 0.15 mm, indicating the effectiveness of the proposed method. The approach provides a practical solution to the inherent trade-off between optical aperture and measurement accuracy, and enables high-precision measurement of large-scale workpieces in confined spaces or with conventional optical equipment.
Keywords:
cross-view field
enhanced vision
transfer calibration
image processing
global measurement
Journal
IF:
3.4
Papers:
2.6K
Citations:
2.3W
Organization
Cited Papers