AI-supported quality control: Flexible, intuitive inspection system for visual inspection processes

Flexible AI inspection

Quality control is under more pressure today than ever before: It must work flexibly, precisely, and economically. Industries such as automotive, mechanical engineering, and electronics manufacturing test a growing variety of components - often under changing conditions and with minimal time for complex integrations. This is precisely where preML's inspection solution comes in. It can be quickly integrated, is intuitive to use, and enables even users without image processing knowledge to create AI-supported inspection models. Their modular design has proven itself in electronic assemblies, plastic parts, and many other industrial applications. The system is delivered as a complete, modular kit - with camera, lighting, edge computing, and preinstalled Software. This makes it a true plug-and-play solution for industrial quality control.

At the heart of the system is the IDS uEye XC industrial camera. Its 13-megapixel sensor delivers detailed images, while the autofocus ensures consistent sharpness - even with changing component heights, variable working distances, or different surfaces.

Practical inspection – from plastic parts to circuit boards

Electronics manufacturing provides a practical example: TOP. Thomas Preuhs GmbH uses the flexible inspection system to check manually assembled printed circuit boards. The combination of powerful camera technology and an AI-based inspection model opens up a new level of automation with minimal operating effort. David Fehrenbach, founder of preML, emphasizes its ease of use: “Our solution fills a crucial gap in the market. For the first time, skilled personnel without knowledge of image processing or AI can independently, intuitively, and directly on the device teach, manage, and perform visual quality inspections.”

uEye XC – Image quality and ease of use

The uEye XC combines the ease of use of a webcam with the reliability of an industrial camera. A single cable connection is all that is needed for installation and commissioning - an advantage in manufacturing environments where systems are frequently converted or expanded. Inside, the 13 MP sensor AR1335 from onsemi delivers high-resolution images for reliable defect detection. The autofocus sharpens automatically and maintains consistent image quality, even with varying object distances.

Jürgen Hejna, Product Manager 2D Machine Vision at IDS, describes it as follows: “The uEye XC is designed for users who require reliable image quality without significant integration effort. The autofocus camera automatically focuses and enables perfect distance adjustment. Its sensor delivers 20 fps at full resolution and, thanks to BSI technology, can also cope with changing lighting conditions.” A quick-change macro attachment lens extends its use at close range: It reduces the minimum object distance and enables the precise detection of small structures, from conductor tracks to material defects and geometric deviations. Features such as digital zoom, auto white balance, and color correction support reliable detail checks.

uEye XC autofocus camera and edge components

September 23-24, 2026 | Santa Clara, CA

Register

“Depending on the inspection requirements, up to four uEye XC cameras can be used per system to map multi-sided inspection processes, complex geometries, or combined surface and detail inspections,” explains David Fehrenbach. “The uEye XC fits perfectly with our system philosophy: It automatically delivers images without our users having to struggle with optical parameters.”

Software & Edge Computing – No code, self-sufficient, ready to use in minutes

The system is delivered with a compact edge computing unit. It operates completely independently, requires no internet connection, and is therefore ideal for isolated production environments. The preinstalled preML Vision Lab software runs as a web application directly on the device. It is designed as a no-code solution and allows test models to be taught and executed without any programming knowledge. Models can be adapted to new products in just a few minutes, which saves time, especially in production with many variants.

Automated quality control of a plastic dowel with the IDS uEye XC and preML Vision Lab: Test model and presentation of results in live view.

Use at TOP: Quality control without expert knowledge

TOP GmbH & Co. KG uses the system to inspect manually assembled printed circuit boards - an environment in which a wide variety of variants, different surfaces, and changing lighting conditions are common. “For over two decades, TOP has been assembling components and end devices of all kinds, from prototypes to small and large series. Production follows the principles of lean production. Many assemblies are created by hand, followed by inductive soldering. Because TOP stands for the highest quality, all manufactured parts undergo a 100 percent inspection. To this end, the company uses the preML GmbH system, among other things,” explains Emil Kurowski, TOP Managing Director.

The solution can be adapted to new products in just a few minutes; test models can be created and expanded without any programming knowledge. Fehrenbach explains: “Our solution combines the best of both worlds: the performance of turnkey inspection systems with the flexibility of customizable software functions. At the same time, it impresses with the operating logic and cost-effectiveness of a smart camera.”

Market development: Flexibility, cost-effectiveness, independence

“The market is clearly moving toward greater flexibility and independence—away from heavy dependence on integrators,” says Fehrenbach. Many manufacturing companies are clearly feeling the effects of this development: They want systems that can be customized without external specialists, can be quickly converted, and can be operated economically. At the same time, price pressure is increasing due to a growing variety of available solutions, making short implementation times and low integration costs increasingly important. The modular preML system addresses these requirements and offers a correspondingly high degree of adaptability.

The system thus opens up new possibilities not only for today's testing processes. It also lays the foundation for future, even more flexible inspection solutions, ranging from automated series testing to fully self-learning quality control. The next step in development is integration into robotics and automation solutions in order to further network testing processes and fully automate them.