Share
Key features
Adoption of an algorithm that combines rule-based defect detection and AI-based defect discrimination
We have more than 10 years of experience in the development and introduction of coating visual inspection equipment. In addition, it has been adopted in many mass-production lines for automotive parts, and has accumulated detection know-how to suppress over-detection.
By utilizing a rule-based defect detection algorithm, it is not necessary to accumulate a large amount of training data. In addition, it avoids detection errors that cannot be analyzed, and then uses AI (machine learning) to identify defective species with high accuracy.
The hybrid configuration of a fixed side and a robot on the top realizes optimal placement of cameras and lighting. Stable images and highly accurate inspections are possible.
Since fixed types are used on the side with few undulations and robot types are used on the upper surface with a large difference in height, imaging and inspection can be performed while following the shape, and high-precision inspection over a wide inspection range is possible.
Classification of defective types and data storage (optional)
It is easy to record and analyze the occurrence and analysis of defect situations that were not easily done by humans until now.
Adaptable to production line speeds of 170 mm/sec (42 s/car, 85 units/h) as standard
※In the case of a 7m pitch (vehicle 5m + distance 2m)
Since the side unit (fixed type) and the top unit (robot type) are separated individually, it is possible to install each individual unit.
After confirming the performance on the mass production line with only the side unit, we can proceed to full-scale introduction.
Defective parts are projected directly onto the vehicle, reducing the load on workers and improving inspection quality.
Advantages of on-vehicle projector marking
Example of use in a process (hybrid display)
