Precise Surface Inspection With AI-Powered Camera Network
In areas such as high-precision metalworking, accuracy is non-negotiable. This includes the automotive industry and its suppliers, which is in crisis in Germany for a variety of reasons. The industry is under enormous pressure to deliver products of impeccable quality and at the same time reduce operating costs, for example by increasing process speeds. The automotive industry’s quality assurance standards are strict, the component geometries are complex and the demands on precision and process reliability are enormous. Fluctuations in detection, insufficient traceability to individual serial numbers or batches and changing environmental conditions can lead to costly errors, production downtime and complaints.
Vivaldi Digital Solutions GmbH has developed an innovative solution for AI quality inspection in real time. In addition to an edge server with an Intel processor, intelligent image processing plays a key role in the so-called SensorBox. Designed for an automotive supplier and its production site in the Czech Republic, up to 70 USB3 uEye+ XC autofocus cameras from IDS Imaging Development Systems perform automatic visual inspection for surface inspection in several inspection cells.
“The IDS cameras integrated into the test cells capture high-resolution images of the surface of aluminium components from German OEMs, in this case an electric motor,” says Uwe Siegwart, Managing Director at Vivaldi, explaining the camera’s task. The captured images are pre-processed to optimise factors such as exposure, white balance and colour correction. The AI recognises various surface defects such as scratches, pores or dirt. “There are strict requirements at the plant in the Czech Republic. According to the defect catalogue, defects from 0.2 millimetres must be reliably detected. The test time per component must not exceed 36 seconds,” says Uwe Siegwart, explaining the stringent guidelines. Up to 2,400 electric motor housings per day are tested 100 per cent automatically with the successfully implemented system.
“The BSI pixel technology enables the sensor to deliver consistently precise and reliable results, even in changing light conditions,” explains Jürgen Hejna, Product Owner 2D cameras at IDS. It is not only the excellent low-light performance and minimal pixel noise that make the sensor particularly suitable for this application. “The 24x digital zoom, combined with auto white balance and precise colour correction, contributes significantly to the success of this demanding test procedure,” emphasises Hejna.
AI-Supported Analyses and Synchronous Camera Control Via Edge Server
The simultaneous image processing task is complex and the challenges are diverse. Aluminium parts often have very shiny surfaces that reflect light and therefore have a strong influence on images. These reflections make it difficult to reliably recognise defects such as scratches, cold runs, chatter marks or surface irregularities, as they are often obscured or distorted by light reflections. This, along with other material and environmental variations, means AI-based image processing must therefore be able to distinguish precisely between the normal material structure and actual raw areas in surfaces to be processed that are considered defects.
Image processing is carried out via an optimised infrastructure. Up to eight cameras can be connected per USB hub, enabling simultaneous recording of several inspection areas. A maximum of two of these USB hubs are connected to a self-developed ‘cam controller’, which bundles and processes the image signals. Several cam controllers are networked with the edge server. This server not only calculates the AI-supported analyses, but also communicates directly with the system controller, which coordinates the triggering of the cameras and ensures synchronised recording.
“Image processing is a central component for traceability, which is the actual technological centrepiece of the solution, and IDS is the ideal partner for this point”, underlines Uwe Siegwart. “Digitised quality data acquisition in real time and linking with other production parameters not only increases efficiency, but also significantly reduces rejects and rework. Errors are minimised, costs are reduced and production processes are optimised.” The system has already paid off in the Czech Republic: The number of complaints per year has been reduced by two thirds and the number of unrecognised surface defects has even been reduced to zero.
The market for digital quality control is developing rapidly, particularly due to more favourable camera technology, for example with regard to higher resolutions. Optimised image processing processes and tools also make it possible for real-time analyses to be carried out by end users,” states Eike Humpert from Vivaldi’s Business Development department. “Improved resolution, more compact design and price-optimised cameras are in demand.”
For more information: www.ids-imaging.com
