Towards automatic testing and defect recognition in magnetic particle inspection
In the trend toward improvement of quality assurance, billions of parts per year are nondestructively tested for surface cracks in the automotive industry. This puts a lot of stress on both the automotive components manufacturing process and the human inspectors. In order to streamline the entire au...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/2016/ http://irep.iium.edu.my/2016/1/FKM-45.pdf |
| Summary: | In the trend toward improvement of quality assurance, billions of parts per year are nondestructively tested for surface cracks in the automotive industry. This puts a lot of stress on both the automotive components manufacturing process and the human inspectors. In order to streamline the entire automotive components production system and to relieve inspector stress, it is necessary to inspect the components automatically based on computer vision systems. This research is intended to tackle the problem of inspecting defects using an existing MPI machine that previously required entirely manual operation. A C++ algorithm has been proposed. It includes special tools that simplify the development of large applications as well as specific libraries that improve productivity. The developed system is much faster and more effective because it became a real-time system. In this system there are two subsystems, the software and the hardware. The hardware consists of an automation system that is responsible for movement of the camera, rotation of the tested work pieces and application of the magnetic particles. All of this mechanism will be controlled in automatic sequences by the software. The software system consists of the image processing algorithm and the control software which controls all the hardware movements. |
|---|