Machine vision solutions can significantly improve quality control in automated manufacturing, increase overall equipment effectiveness, and reduce costs, such as for end-of-line quality control in food production. However, developing robust applications is challenging, and standalone solutions using smart cameras or industrial PCs are expensive to maintain and difficult to scale. That’s why Siemens is collaborating with Basler and MVTec in its open Industrial Edge Ecosystem. The results are scalable, plug-and-play solutions that combine Siemens’ automation technology with third-party machine vision hardware and software. Under this new partnership, Basler and MVTec have developed innovative applications that run on Siemens’ Industrial Edge platform, which combines central device management with de-centralized data processing in factories.

Machine vision solutions can significantly improve quality control in automated manufacturing. (Image: Siemens AG)

Based in Ahrensburg, Germany, Basler manufactures industrial cameras, accessories and components that are used in factory automation, medicine, robotics and a wide variety of other industry verticals. The most common applications involving machine vision include defect detection, positioning and measuring parts, and identifying, sorting and tracking products. Their machine vision cameras are already involved in the type of industrial applications that the partnership’s Industrial Edge Ecosystem is targeting. One recent example involves a segment of Germany’s national railway system.

The Deutsche Bahn, Germany’s national railway company, developed a machine vision process using Basler’s cameras in a 32-camera “arch” arrangement for visual inspection of train surfaces for maintenance needs. Trains entering the Deutsche Bahn maintenance hall pass through the camera arch, which records two high-resolution images per second of the trains. Mobile E-Check cobots are then used in the maintenance hall. Each cobot is equipped with two robotic arms and one Basler ace 2 camera per arm. They take care of areas that cannot be captured by static cameras. The data collected by the cameras is processed using the “AutomateOS” middleware from Gestalt Robotics and made available for subsequent steps. A sophisticated AI system then detects and classifies any anomalies and defects. Examples include the detection of missing screws, material wear and tear, or damage caused by stone chips.

Industrial Edge-based visual inspection of battery cells with MVTec Merlic (left) and Simatic connectivity solutions with Basler cameras and Basler Vision Connector App (right). (Image: Siemens AG)

Connecting the dots between the physical cameras, robotics and infrastructure with the automated software, AI algorithms and models that make this type of machine vision possible in an easy plug-and-play approach is exactly what the Industrial Edge Ecosystem has been established for. It is easy to learn, test, and use because it does not require programming or an in-depth knowledge of machine vision.

The Basler ‘Vision Connector’ app connects Basler’s industrial cameras to the Siemens automation portfolio, allowing camera parameters to be customized for specific use cases and providing the flexibility to create and deploy custom vision apps or artificial intelligence (AI) models. MVTec’s ‘Anomaly Detection for Visual Inspection’ technology provides camera connectivity and image analysis in a single application that uses two deep learning networks for defect detection, eliminating the need for programming or deep vision expertise. Siemens′ partnerships with Basler and MVTec allow industrial companies to focus on their specific use cases without spending excessive time and resources on technology integration.

Quality control is critical in modern industry. Machine vision makes it less error-prone, time-consuming, and costly. By adopting an end-to-end platform like the Industrial Edge ecosystem, new scalable machine vision solutions can be efficiently and seamlessly integrated into production automation.

Digitalization and automation are key drivers for manufacturing competitiveness. Machine vision technologies can provide significant support and contribute to improved quality control. The result leads to improved overall equipment effectiveness, higher quality and transparency, and lower costs, for example in end-of-line quality control at a multinational consumer packaged goods manufacturer. But before these opportunities can be realized, industrial manufacturers must overcome a number of challenges. For example, developing robust applications with vision algorithms and adapting and integrating them to the production environment and conditions, which requires a high level of expertise. Or avoiding stand-alone solutions with smart cameras or industrial PCs that generate significant system maintenance efforts.

Architecture of the machine vision solution with Siemens Industrial Edge and machine vision apps from Basler or MVTec. (Image: Siemens AG)

Industrial Edge Ecosystem

So how can isolated solutions be avoided and machine vision applications be integrated into the automation landscape in a way that is scalable? The Industrial Edge platform from Siemens combines central device management with decentral, machine-oriented data processing — also known as edge computing. With the Industrial Edge ecosystem, Siemens is pursuing an open approach in which industry-proven solutions from third-party vendors and Siemens can be combined and seamlessly integrated into automation systems. Industrial manufacturers and machine builders benefit from the compatibility with the Siemens Simatic portfolio and the offerings among each other.

In the Industrial Edge Marketplace, an application store for the industry, customers can select the most appropriate software components to create flexible machine vision solutions for their specific use cases. This makes isolated solutions a thing of the past: camera-software combinations can be securely, easily and integrated with a scalable architecture into the automation setup on any number of edge devices and updated with just a few mouse clicks via the Industrial Edge Management System. The cooperation between Siemens, Basler and MVTec makes these benefits a reality with scalable, plug-and-play solutions that combine Siemens’ automation technology with third-party machine vision hardware and software.

Basler and MVTec have launched first Apps

Basler and Siemens have developed a new joint interface to connect Basler industrial cameras to the Siemens automation portfolio. The first Industrial Edge application based on this interface is the ‘Basler Vision Connector’, which allows the camera parameters to be adapted to individual requirements of a specific use case. A future Simatic technology module will use the same interface as the Edge app to enable a direct connection between Basler cameras and the S7-1500 system. Compared to other solutions, the open approach of Siemens and Basler offers the flexibility to develop and use a manufacturer’s own individual vision applications or AI models. The ‘Anomaly Detection for Visual Inspection’ application developed by MVTec is also part of the Industrial Edge ecosystem. It provides camera connectivity and image analysis in a single app, without the need for programming or in-depth vision expertise. It uses two deep learning networks to detect not only defects such as scratches or dirt, but also missing or misplaced components.

Complexity Down, Optimization Up

With the described hardware-software solutions on Industrial Edge and the Simatic S7-1500 system, industrial image processing is not being reinvented, but rather orchestrated differently and tailored to the respective customer requirements. Integrating the camera and software into automation systems makes it easy to access and analyze data to optimize production processes and ensure product quality. Scalability through Industrial Edge means that industrial vision use cases can be rolled out in a centrally managed way. This approach of a standardized, scalable machine vision infrastructure allows industrial companies to focus on the specific challenges of their use cases without having to spend too much time and resources on technology integration. With Siemens Industrial Edge and machine vision offerings from Basler or MVTec, this is now possible, ensuring seamless integration into new and existing automation systems.

This article was written by, Stefan Meier, Ecosystem Manager Industrial Edge Solutions at Siemens (Nuremberg, Germany). For more information, visit here  .