Image-processing system for circular-grid analysis in sheet-metal forming

1994 ◽  
Vol 34 (2) ◽  
pp. 108-115 ◽  
Author(s):  
R. -S. Lee ◽  
Q. -C. Hsu
2000 ◽  
Vol 40 (2) ◽  
pp. 190-196 ◽  
Author(s):  
C. -C. Wang ◽  
J. Lee ◽  
L. -W. Chen ◽  
H. -Y. Lai

Author(s):  
Wankhede Pankaj ◽  
Tejas Radhakrishnan ◽  
Kurra Suresh ◽  
Sudha Radhika

Strain analysis is an essential step in sheet forming applications to understand the deformation behaviour of the material. The strain analysis is conventionally performed by Circle Grid Analysis (CGA). In CGA, a grid of circles with 2 mm–5 mm diameter is printed on a flat sheet and deformed into the required shape. The length of major and minor axes of the deformed ellipses are measured to estimate the strains at different locations of the formed component. The major and minor axes length of the deformed ellipses are measured either by using a manual method or automatic methods. The automated methods are faster and accurate compared to manual methods, but to perform strain measurement using commercially available automatic strain measurement system, is very expensive. Therefore, in this paper, an image processing based software has been developed with user-friendly Graphical User Interface (GUI) for strain measurement and analysis in sheet metal forming operations. The software has been tested on images obtained by printing the ellipses of known dimensions on the flat sheet by laser etching, electrochemical etching and screen printing, which are generally used for printing circular grids on flat sheets for strain analysis in sheet metal forming. It is observed that the edges of the laser engraved ellipse are very slim and easily detected by software compared to other grid printing methods. The developed software can measure the ellipse dimensions with a maximum absolute percentage error of 1.975%. Further, the results of CGA software has been compared with the commercial strain measurement system called Grid Pattern Analyzer (GPA). The mean absolute error in strain measurement using CGA and GPA was found to be 0.0106 and 0.0178, respectively. The statistical test results reveal that there is no evidence to support a claim that there is a difference in mean performance between the two methods.


Sign in / Sign up

Export Citation Format

Share Document