A coordinated push by international standards bodies is enabling machine vision systems from competing vendors to share real-time defect data across high-mix metal fabrication lines-a shift that integrators and plant managers say has moved beyond controlled pilots onto active production floors.
Background
High-mix, low-volume fabrication environments have historically resisted automated vision integration because no single vendor's inspection hardware covers every process step. Laser cutting, press brake forming, welding, and finishing each carry distinct sensing requirements. The result: fragmented data, manual handoffs between inspection stations, and quality escapes tied to part-to-part variation.
The standards groundwork enabling cross-vendor integration has accumulated over years. In 2009, the Association for Advancing Automation (A3), the European Machine Vision Association (EMVA), and the Japan Industrial Imaging Association (JIIA) entered a cooperation agreement on machine vision standards, forming the "G3" coalition. Germany's VDMA and China's CMVU joined in 2014 and 2015, respectively. The coalition coordinates standards including GenICam, GigE Vision, and USB3 Vision-the transport and interface protocols that underpin multi-vendor compatibility at the hardware level.
The governance layer for shop-floor data exchange arrived separately. The OPC UA companion specification for machine vision targets straightforward integration of vision systems into production control and IT infrastructure-not merely replacing existing point-to-point interfaces, but creating new horizontal and vertical integration capabilities between authorized process participants. The OPC UA Vision interface can exchange information between a machine vision system and another vision system, a machine PLC, a line PLC, or any software system at the control device level.
Details
Adoption of these protocols at the production level is now measurable. A 2025 VDMA study found that 57 percent of companies use OPC UA in production, confirming the technology has left the pilot phase and become industrial reality, driven by customer demands and strategic efficiency pressure. The same study found that 84 percent of companies see a concrete need for interoperable interfaces, with 71 percent rating OPC UA as highly relevant and 62 percent considering companion specifications particularly important.
For fabricators operating mixed-vendor lines, the practical effect is that real-time inspection results-weld seam geometry, edge straightness, surface anomaly classifications-can now be published to a shared data bus that PLCs, MES platforms, and ERP systems each consume without custom middleware. Unified standards counter dependency on proprietary solutions and prove particularly beneficial for machine vision applications, where many different components interact, reducing complexity and increasing process efficiency.
On the physical transport layer, GigE Vision 3.0 is advancing support for RDMA over Converged Ethernet (RoCEv2) to enable devices to transfer image data without involving the host operating system, according to A3, reducing latency on high-throughput inspection cells. EMVA's sensor quality standard, ISO 24942 (EMVA 1288), was promoted to ISO level in October 2024 and has been described as the first machine vision standard accepted as an international standard-giving procurement engineers a vendor-neutral benchmark for qualifying cameras across a multi-supplier fleet.
Cybersecurity has emerged as a non-trivial implementation barrier. As machine vision systems grow more connected, new standards are being developed to protect them from cyberattacks and ensure data integrity and system performance. The ISA/IEC 62443 series defines requirements and processes for implementing and maintaining electronically secure industrial automation and control systems, setting best practices for security and providing a framework to assess security performance levels. Fabricators integrating cross-vendor vision networks face increasing pressure to demonstrate zone segmentation and access control conformance-particularly when inspection data flows to cloud-hosted MES or enterprise analytics platforms.
Implementation also carries real integration overhead. The ongoing lack of interoperable camera API standards results in unnecessarily high integration costs for new camera technologies, increasing application development time and maintenance costs while reducing software portability.1IEC 62443 Standard: Enhancing Cybersecurity for Industrial Automation and Control Systems | Fortinet The Khronos Group's Kamaros Working Group, established in 2022 alongside EMVA, is developing an open, cross-vendor embedded camera API standard to address that gap at the sensor control layer-though that work remains in progress.
Outlook
Roughly half of new industrial products are already OPC UA-capable, signaling broad acceptance and a clear transition from pilot phase to real implementation. Standards bodies are expected to advance companion specifications for specific fabrication sub-processes-including welding inspection and dimensional gauging-over the next 12 to 24 months, which would lower integration effort for shops adopting multi-vendor vision on press brake and laser cutting cells. Fabricators that standardize on OPC UA Vision and IEC 62443-aligned security architectures now position themselves to onboard next-generation AI-driven inspection modules without rearchitecting their data infrastructure.
