1306 Real-time Visual Monitoring of Welding Process in MAG Welding
Junya Ninomiya, Kazuki ICHIKAWA, Akiyoshi Miyamoto, Yasuo Suga
Abstract
Open-access reader
Junya Ninomiya, Kazuki ICHIKAWA, Akiyoshi Miyamoto, Yasuo Suga
Abstract
Open-access reader
The automatic welding is widely used for the production of machines and assembly of constructions. It will increase the efficiency of a welding process, productivity and quality of welded joints, if the ability of welding robot and sensors which can monitor the welding process increases. However, there are two problems that the arc welding process is unstable welding and spatters occur in MAG welding and they disturb the vision sensing of welding process. A purpose of this research is construction of a system which is possible to obtain sound welding of joints in the automatic MAG welding. This system monitors the welding area by CCD camera and the images are processed. Then the system detects welding wire and welding line position, and controls the welding robot to realize the automatic seam-tracking. The new image processing algorithm was constructed to eliminate the bad influence of spatters on a welding image and increase the successful percentage of the welding line detection and the welding wire detection.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The automatic welding is widely used for the production of machines and assembly of constructions. It will increase the efficiency of a welding process, productivity and quality of welded joints, if the ability of welding robot and sensors which can monitor the welding process increases. However, there are two problems that the arc welding process is unstable welding and spatters occur in MAG welding and they disturb the vision sensing of welding process. A purpose of this research is construction of a system which is possible to obtain sound welding of joints in the automatic MAG welding. This system monitors the welding area by CCD camera and the images are processed. Then the system detects welding wire and welding line position, and controls the welding robot to realize the automatic seam-tracking. The new image processing algorithm was constructed to eliminate the bad influence of spatters on a welding image and increase the successful percentage of the welding line detection and the welding wire detection.
Key concepts: Welding, Robot welding, Arc welding, Process (computing), Welding power supply, Electrogas welding, Mechanical engineering, Machine vision