Share Email Print

Proceedings Paper

A distortion-correction method for workshop machine vision measurement system
Author(s): Ruwen Chen; Ren Huang; Zhisheng Zhang; Jinfei Shi; Zixin Chen
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The application of machine vision measurement system is developing rapidly in industry for its non-contact, high speed, and automation characteristics. However, there are nonlinear distortions in the images which are vital to measuring precision, for the object dimensions are determined by the image properties. People are interested in this problem and put forward some physical model based correction methods which are widely applied in engineering. However, these methods are difficult to be realized in workshop for the images are non-repetitive interfered by the coupled dynamic factors, which means the real imaging is a stochastic process. A new nonlinear distortion correction method based on a VNAR model (Volterra series based nonlinear auto-regressive time series model) is proposed to describe the distorted image edge series. The model parameter vectors are achieved by the laws of data. The distortion-free edges are obtained after model filtering and the image dimensions are transformed to measuring dimensions. Experimental results show that the method is reliable and can be applied to engineering.

Paper Details

Date Published: 31 December 2008
PDF: 7 pages
Proc. SPIE 7130, Fourth International Symposium on Precision Mechanical Measurements, 71304B (31 December 2008); doi: 10.1117/12.819715
Show Author Affiliations
Ruwen Chen, Southeast Univ. (China)
Nanjing Institute of Technology (China)
Ren Huang, Southeast Univ. (China)
Zhisheng Zhang, Southeast Univ. (China)
Jinfei Shi, Southeast Univ. (China)
Zixin Chen, Southeast Univ. (China)

Published in SPIE Proceedings Vol. 7130:
Fourth International Symposium on Precision Mechanical Measurements
Yetai Fei; Kuang-Chao Fan; Rongsheng Lu, Editor(s)

© SPIE. Terms of Use
Back to Top