A consortium of North American Tier 1 metal suppliers has moved cross-vendor machine vision standards from controlled testing into live production trials, marking a significant step toward interoperable automated inspection across cutting, punching, and welding cells.
The initiative standardizes image data formats, AI model interfaces, and communication protocols across multiple vision platform vendors. It is now running in active production environments at Tier 1 fabrication operations. Early trial reports indicate smoother system integration, reduced deployment downtime, and cleaner data pipelines feeding quality assurance and regulatory reporting functions.
Background
Fragmented vision infrastructure has long complicated quality inspection in high-mix metal fabrication. When a production floor runs laser cutting, CNC punching, and robotic welding cells from different equipment vendors, each subsystem historically generated inspection data in proprietary formats - creating bottlenecks for MES integration and hindering real-time tolerance monitoring across process handoffs.
The standards framework underpinning these trials draws on established cross-vendor specifications. The VDMA OPC Vision Initiative is a joint effort of VDMA Machine Vision and the OPC Foundation, aimed at developing an OPC UA companion specification for machine vision. Launched in January 2016, the initiative released Part 1 of the OPC Machine Vision Companion Specification in 2019 and Part 2 in 2024. The companion specification targets straightforward integration of machine vision systems into production control and IT systems - not only complementing existing interfaces but enabling horizontal and vertical integration up to the enterprise IT level.
Alongside OPC UA for Machine Vision, the trials incorporate GigE Vision and GenICam protocols. Multiple global machine vision organizations - A3, EMVA, JIIA, and VDMA - cooperate to avoid duplication, provide education, and coordinate interoperability events, as presented at Automate 2025. Separately, Werner Feith of the European Machine Vision Association noted that ISO-24942 (EMVA 1288) measures camera and image sensor quality. EMVA 1288 was promoted to ISO level in October 2024, described as the first machine vision standard accepted as an international standard.
Details
The live plant trials target a specific operational gap: the absence of a unified data layer across multi-vendor inspection nodes positioned at sequential fabrication stages. The OPC UA infrastructure layer enables simplified, uniform integration of machine vision systems into higher-level IT production systems - PLC, SCADA, MES, ERP, and cloud. Subsequent specification parts are intended to replace proprietary elements with standardized information structures and semantics, including configuration, recipe, and result data.
The specification allows vision system functionalities to be addressed manufacturer-independently and integrated on the IT side. Hardware demonstrators have shown the simultaneous interaction of multiple clients with image processing systems from different suppliers.
The trials unfold against a broader backdrop of accelerating inspection automation. The global machine vision market is expected to grow from USD 15.83 billion in 2025 to USD 23.63 billion by 2030, at a CAGR of 8.3%, according to MarketsandMarkets. The fastest-growing segment is AI-based software, which serves as the intelligence layer enabling variant-flexible inspection and robotic guidance without hard-coded programming.
North American Tier 1 fabricators are advancing AI vision-guided automation into live production as OPC UA standards and SBOM governance shape deployment outcomes. On the cybersecurity front, procurement teams at larger manufacturers now require Software Bills of Materials (SBOM) from vision platform vendors1Automatic Visual Inspection for Industrial Application - PMC - a compliance layer the consortium is incorporating into its data governance framework for the trials.
A critical interoperability gap remains under active remediation. Standardization bodies are expected to address cross-vendor telemetry and data-format compatibility over the next 18 to 24 months - a gap that currently complicates factory-wide rollouts involving multiple vision platform suppliers.22026 Guide to Automated Visual Inspection Systems Manufacturers that resolve ERP integration and OT segmentation early stand to compress subsequent site deployments from weeks to days.3Automated Visual Inspection: — Proven Strategies That Actual
According to the Manufacturing Leadership Council's 2025 "Shaping the AI-Powered Factory of the Future" report, 72% of manufacturers surveyed said they are currently using AI solutions in the form of a vision system. While these systems can enhance operational efficiency and reduce costs, human inspectors continue to play a critical role overseeing manufacturing quality control alongside automated systems.
Outlook
U.S. metal fabricators are scaling AI defect detection to full production, with federated AI inspection models sharing anonymized updates across facilities to boost accuracy without exposing proprietary data. Active standards development continues across Camera Link HS, GigE Vision, USB3 Vision, and GenICam, with CoaXPress v3.0 targeting 25 Gbps transmission and advancing through product validation. For Tier 1 suppliers completing these trials, a successful outcome would establish a replicable deployment template - accelerating rollout to additional production lines and supplier tiers across the North American metalworking supply chain.
