Share Email Print
cover

Proceedings Paper

FPGA and DSP based an intelligent visual sensor design for laser welding seam recognition
Author(s): Fang Jiang; Chunying Jiang; Hanlin Ge
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

A kind of intelligent visual sensor system, which is mainly composed of a laser lighting source and an intelligent camera based on FPGA and DSP, is designed using the triangle structured light principle for laser welding seam recognition. Intending to meet the high-precision demands, on one hand, the components of the sensor such as the lighting source, the optical lens, the optic sensor of the camera are selected and the errors caused by these parts are analyzed. Furthermore, the triangle structured light principle models are built and the simulation is given to optimize the light path parameters. On the other hand, the methods of the image processing for seam recognition using intelligent camera based on FPGA and DSP are explained. Experiments results proved that the sensor design is reasonable and the image recognition methods are rational and effective. Very compact structure, high-precision and high-real-time performance are integrated in this sensor, so it can be applied well in the seam tracking and inspection system in the laser welding and arc welding fields.

Paper Details

Date Published: 12 October 2010
PDF: 7 pages
Proc. SPIE 7656, 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment, 765677 (12 October 2010); doi: 10.1117/12.864526
Show Author Affiliations
Fang Jiang, Liaoning Shihua Univ. (China)
Chunying Jiang, Shenyang Institute of Automation (China)
Hanlin Ge, Liaoning Shihua Univ. (China)


Published in SPIE Proceedings Vol. 7656:
5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment
Yudong Zhang; Jose M. Sasian; Libin Xiang; Sandy To, Editor(s)

© SPIE. Terms of Use
Back to Top