Paper
13 June 2024 Triangle workpiece precision dimensional inspection
Jiahao Yu, Jiacheng Liu, Liankai Wang, Zhan Yang, De Song
Author Affiliations +
Proceedings Volume 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024); 131803X (2024) https://doi.org/10.1117/12.3033756
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 2024, Guangzhou, China
Abstract
In response to the difficulties in measuring the dimensions of triangular precision workpieces and the high cost of inspection, this paper proposes a fast and accurate detection algorithm based on partial-area-effect model. Firstly, image preprocessing is performed through morphological operations. Then, the OTSU adaptive threshold is introduced into the partial-area-effect model algorithm to improve detection accuracy. Afterwards, straight edge information is extracted using 8-connected region Hough transform. Finally, the improved least squares method combining least squares with RANSAC method is used to fit the length of the straight line. This method has fast computation speed, high detection accuracy, with maximum measurement error less than 0.004mm, average measurement error accuracy less than 0.003mm, and average detection time less than 1.5s. It meets the accuracy requirements for workpiece inspection.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Jiahao Yu, Jiacheng Liu, Liankai Wang, Zhan Yang, and De Song "Triangle workpiece precision dimensional inspection", Proc. SPIE 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 131803X (13 June 2024); https://doi.org/10.1117/12.3033756
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Edge detection

Image processing

Detection and tracking algorithms

Hough transforms

Inspection

Image processing algorithms and systems

Image segmentation

RELATED CONTENT


Back to Top