Camera recognises & classifies demanding surface defects

9th July 2014

Staff Reporter

0 0

A surface inspection system, which can be used in all stages of the steel strip production, has been developed using the RazerCam smart camera series. The system, developed by EVT, can be applied in real time due to the free-programmable FPGAs of the camera complex filters. Two different systems are available, which are either equipped with a line scan camera or with a matrix camera. A combination of both camera types is also possible.

Crucial errors on the entire steel strip surface can be detected by the user, with the system then able to classify the errors depending on shape and size according to the production titles such as roller dents, reel slide, etc. With the EyeVision evaluation software the system flexibly variable for the automated steel strip inspection during all process sections – from the hot-rolled strip to the cold roll. Additionally, thanks to the 2 ARM Cortex A9 cores, the RazerCam is applicable for every production speed.

The illumination of steel has a great influence on the detection of defects on steel strips - users have the choice between bright field and dark field illumination. With a bright field illumination the errors, which reflect or disperse light, are displayed dark on a bright background. Especially defects such as scratches and dents are easily recognizable in this light. With a dark field illumination the errors are displayed bright on dark background. But as images in dark field illumination have to be amplified due to their low light quantity, the contrast is not as good as with bright field illuminations.

In short, the RazerCam offers the possibility to recognize and classify demanding surface defects. Thanks to the FPGAs of the RazerCam, the image inspection is done in real time and therefore is faster than to date.