We have developed an automatic teaching system for laser welding machine. As it is a kind of a robot application, operator have to teach trajectories for each work. After rough manual teaching of start and end points, the system tracks the seam line automatically and finally generates an exact teaching program for the actual welding. Therefore, the operators’ burden for teaching and operating time are decreased by using this system. An image processing method, Hough transform, is applied for seam line detection. This algorithm processes an image taken by a CCD camera on the laser torch and extracts the position and angle of the seam line. Although Hough transform requires long processing time, we have made a hardware circuit for Hough transform to realize real-time seam tracking within 0.4 seconds per image.

Topics
Charge coupled devices, Image processing, Robotics, Machining, Welding
1.
G
Cook
, (1983).
Through-the-arc sensing: Positioning control for robotic arc welding
,
Proc 1st Annu Int Rob Conf
1983
, pp.
79
–
94
.
Google Scholar
 
2.
J
Nakajima
, et al, (
1982
).
Development of semiautomatic arc welder controlled by microcomputer
,
Fundam Pract Approaches Reliab Welded Struct
vol.
1
, pp.
217
–
222
.
Google Scholar
 
3.
U
Heitmayer
 and
K
Weman
, (
1981
).
Adaptive sensing: key to the adult robot
,
Weld Met Fabr
, vol.
49
No.
10
, pp.
592
–
596
.
Google Scholar
 
4.
N
Kodaira
, et al, (
1982
).
Microcomputer control of an arc welding robot with visual sensor
,
IEEE 1982 IECON Prof
, pp.
191
–
195
.
Google Scholar
 
5.
T
Bamba
, et al, (
1984
).
A visual seam tracking system for arc-welding robots
,
Proc 14th Int Symp Ind Rob 7th Int Conf Ind Rob Technol
, pp.
365
–
374
.
Google Scholar
 
6.
Hough
, (
1962
).
P.V.C. Method and means for recognizing complex patterns
, U.S. Patent 3, 069, 654,
Dec.
18,.
7.
R. O.
Duda
 and
P. E.
Hart
, (
1972
).
Use of the Hough transform to detect lines and curves in pictures
,
Communications of the ACM
, vol.
15
, pp.
11
–
15
.
Google Scholar
Crossref
Search ADS
 
8.
J
Yamato
, et al, (
Jan
,
1989
).
Highly Accurate Segment Detection Using Hough Transform
,
IEICE Trans.D-2
, Vol.
J72-D-2
, No.
1
, pp.
85
–
92
.
Google Scholar
 
9.
H
Koshimizu
, (
Nov.
19,
1987
).
ON THE CURRENT TRENDS OF HOUGH TRANSFORM IN PATTERN PROCESSING
,
Computor Vision
51-1
, pp.
1
–
8
.
Google Scholar
 
10.
H
Koshimizu
,
On the Current Trends of Hough Transform Researches(Part-2
),
TECHNICAL PEPORT OF IEICE.PRU91-15
, pp.
25
–
32
11.
T
Matsuyama
,
H
Koshimizu
, (
Sep
,
1989
).
Hough Transform and Pattern Matching in Computor Vision
,
Journal of IPSJ
vol.
30
No.
9
, pp.
1035
–
1046
.
Google Scholar
 
12.
J
Toriwaki
, (
1988
).
Digital Image Processing for Image Understanding[1]
,
SYOUKOUDOU
,.
Google Scholar
 
13.
J
Toriwaki
, (
1988
).
Digital Image Processing for Image Understanding[2]
,
SYOUKOUDOU
.
Google Scholar
 
14.
H
Tamura
, (
1985
).
Computor Image Processing
,
SOUKEN-SYUPPAN
.
Google Scholar
 
15.
H
Koshimizu
,
M
Numada
, (
1988
)
A Fast Incremental Hough Transform FIHT
,
TECHNICAL PEPORT OF IEICE.PRU97-93
, pp.
1
–
6
.
Google Scholar
 
16.
H
Koshimizu
,
M
Numada
,
FIHT2:A Fast Incremental Hough Transform
,
TECHNICAL PEPORT OF IEICE.PRU88-107
, pp.
7
–
14
.
17.
H
Koshimizu
,
M
Numada
, (
Jan
,
1989
)
On A Fast Hough Transform Method PLHT Based on Piece-Wise Linear Hough Function
,
IEICE Trans.D-2
, vol.
J72-D-2
, No.
1
, pp.
56
–
65
.
Google Scholar
 
This content is only available via PDF.
© 1997 Laser Institute of America.
1997
Laser Institute of America
You do not currently have access to this content.