Assembly quality detection method based on PointNet++ and point set registration

Wanyan Huang; Wanhe Du; Tao Yu

doi:10.1117/12.2645509

30 August 2022 Assembly quality detection method based on PointNet++ and point set registration

Wanyan Huang, Wanhe Du, Tao Yu

Proceedings Volume 12309, International Conference on Advanced Manufacturing Technology and Manufacturing Systems (ICAMTMS 2022); 1230912 (2022) https://doi.org/10.1117/12.2645509
Event: International Conference on Advanced Manufacturing Technology and Manufacturing System (ICAMTMS 2022), 2022, Shijiazhuang, China

Abstract

To solve the problem of missing assembly, a non-contact detection algorithm for assembly quality was proposed based on PointNet++ and point set registration. Firstly, the contour information of the target component was preprocessed using filtering and a local limiting filtering algorithm, and the contour data was sampled down and the contour peaks were processed. Then, the target contour was identified by the PointNet++ network. Finally, the point set registration method was used to register the target contour with the standard point set, and then the desired contour assembly information was obtained. Taking the engine cylinder valve as an example, the experimental results show that this method can effectively identify the assembly quality of the target component.

Citation Download Citation

Wanyan Huang, Wanhe Du, and Tao Yu "Assembly quality detection method based on PointNet++ and point set registration", Proc. SPIE 12309, International Conference on Advanced Manufacturing Technology and Manufacturing Systems (ICAMTMS 2022), 1230912 (30 August 2022); https://doi.org/10.1117/12.2645509

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available