For vision-based measurement, there are few research or professional tools for local contour positional errors of flexible automotive rubber strips. To support the automatic measurement of contour positional errors, a novel local contour registration and measurement method based on shape descriptors is proposed. In this method, a shape descriptor is proposed to find correspondence between a reference local contour and a desired local contour. First, a shape descriptor that includes the shape representation and restrictions of the local contour is extracted from the reference contour. Second, several tolerable shape descriptors for a desired actual local contour are constructed by adding some loosening factors to the ideal descriptor, and an angular similarity-based searching strategy is used to find the best actual local contour. Finally, from the matched local point sets, a quantitative calculation step provides the desired deviation values. This method is implemented in a sealing strip cross section measurement system, and numerous cross-sectional profiles are tested. The experimental results verify the stability and effectiveness of the proposed method. Important progress toward the automatic measurement of flexible products is demonstrated.
Shape Descriptor-Based Local Contour Profile Registration and Measurement for Flexible Automotive Sealing Strips
Manuscript received October 24, 2016; final manuscript received February 7, 2018; published online March 16, 2018. Editor: Satyandra K. Gupta.
- Views Icon Views
- Share Icon Share
- Cite Icon Cite
- Search Site
Li, J., Du, Z., and Wang, Y. (March 16, 2018). "Shape Descriptor-Based Local Contour Profile Registration and Measurement for Flexible Automotive Sealing Strips." ASME. J. Comput. Inf. Sci. Eng. June 2018; 18(2): 021006. https://doi.org/10.1115/1.4039430
Download citation file:
- Ris (Zotero)
- Reference Manager