Deep Learning-based Surface Inspection
Precise surface inspection in battery production effectively identifies scratches and paint accumulations on coated surfaces. Coating recesses are precisely measured and checked for their positioning.
For TAS GmbH, we have developed an automated solution for inspecting surface coatings. Through its precise finishing of metal surfaces, TAS plays a crucial role in the manufacturing of automotive components – especially for the covers of electric vehicle battery compartments, where high-voltage batteries are installed. The finished surfaces protect the high-voltage batteries from external influences such as moisture, as they could otherwise ignite due to corrosion. The battery compartments must therefore be sealed precisely with the covers.
TAS comprehensively and carefully inspects each part for defects. To automate this process, we have utilized state-of-the-art camera technology and advanced machine vision software for detection. For this purpose, a robotic arm feeds each individual battery cover to a testing fixture, where a camera system examines the covers for possible defects such as irregularities in the coating, scratches on the surface, or issues with laser marking.

The unique aspect of this machine vision solution is the Deep Learning technology from Zebra Aurora Vision Studio. It can be trained using images of specific defect types to recognize and classify them. For this purpose, pre-selected image datasets are annotated and supplied to the system. Based on this information, the system can then be retrained for new inspection criteria or fine-tuned for defects already known to it. As a result, the machine vision solution can be continuously developed and improved through Deep Learning technology.
The need for precision and reliability in detecting even the smallest surface defects, which could impair the reliable sealing of electric vehicle battery compartments with their corresponding covers, was crucial for the customized camera system and the machine vision solution.
Success Story
We don’t have customers, we have partners.
I was particularly impressed by the fact that the competent and motivated staff always responded to interim changes to the plan in a friendly, quick, and flexible manner, and that our requested changes were implemented at no additional cost
ID has excellent, motivated, and friendly specialists in manufacturing process automation who know exactly what they're doing. From conception to production, all the experts are in-house, which made working with us much easier.
We supply high-quality components for industrial image processing, which ID uses in special-purpose machine construction. In addition, ID conducts professional training courses for our software at customers' sites, which significantly contributes to the efficient use of the components in automation.

