Quality
Visual quality inspection
Cameras and computer vision models inspect parts on the line to catch surface, assembly and dimensional defects consistently.
Buying guide
Software and systems that use cameras and machine learning to detect defects, verify assembly and measure parts on the production line.
AI-based vision inspection software learns to distinguish good parts from defective ones using labeled images, which helps with defects that vary in appearance and are hard to describe with fixed rules, such as surface scratches, porosity, contamination, weld irregularities and cosmetic flaws. Offerings range from software that runs on your own cameras and industrial PCs, to integrated smart cameras, to complete turnkey stations delivered with lighting, optics and mechanics.
Software is only one part of a working inspection system. Lighting, optics, part presentation, labeling discipline and integration with the PLC often decide the outcome. When evaluating products, look at the whole path from image capture to reject handling, and at how the system will be maintained after the integrator leaves.
From each company's published materials. "Not published" means we could not find it, not that it does not exist.
| Company | Main product | Use cases | Deployment | Pricing | Founded |
|---|---|---|---|---|---|
| Avathon | Knowledge Platform | Predictive maintenance, Process anomaly detection, Visual quality inspection | Not stated | Not published | 2013 |
| Cognex | OneVision | Visual quality inspection | Edge (on-camera or PC); OneVision trains models in the cloud and runs inspection at the edge | Not published | 1981 |
| Cybord | Enforce | Visual quality inspection, Supply chain risk & visibility | Air-gapped deployment option announced (per Cybord press release) | Not published | 2018 |
| Eigen Innovations | OneView Edge | Visual quality inspection, Quality prediction & virtual metrology, Predictive maintenance | Edge + cloud (OneView Edge and OneView Cloud) | Not published | 2012 |
| Elementary | VisionStream | Visual quality inspection | Edge + cloud (edge controllers on the line; QualityOS hosted on AWS) | Not published | 2017 |
| Instrumental | Vision Engine | Visual quality inspection, Root-cause analysis for quality & yield | Not stated | Not published | 2015 |
| LandingAI | LandingLens | Visual quality inspection, Document automation for supplier, compliance & QMS | LandingLens: cloud or edge devices; document APIs: cloud, on-premises or virtual private deployment (per landing.ai) | Not published | 2017 |
| Siemens | Siemens Industrial Copilot | Copilots for PLC & automation engineering, GenAI knowledge assistants & copilots, Predictive maintenance | Not stated | Not published | 1847 |
Training approaches that cope with rare defects, such as learning mainly from good parts or making efficient use of a small number of defect images.
Built-in tools for labeling, review, labeler agreement checks and versioned datasets, with traceability from each model back to the images that trained it.
Inference on hardware at the line that meets your cycle time with margin and keeps running if the network is down.
Support for triggers, results and reject signals over the industrial protocols you use, plus reliable part tracking to the reject point.
Storage of images and decisions linked to serial number, lot or timestamp for quality investigations and customer inquiries.
Versioning, validation against a held-out test set, staged deployment and rollback, compatible with your change control or validation procedures.
Defects are visible with the right imaging, current inspection is manual, inconsistent or a throughput bottleneck, and you can present parts consistently at a station.
The defect is internal or not visible with any practical imaging method, part presentation cannot be controlled, or a simple rule-based vision check already solves the problem reliably.
Quality
Cameras and computer vision models inspect parts on the line to catch surface, assembly and dimensional defects consistently.
Operations & production
Cameras and sensors guide operators step by step and verify each manual assembly step, catching errors before the part moves on.
Quality
Machine learning sifts process, material and equipment data to narrow down which factors are actually driving defects and yield loss.
Put AI to work
Genvo, our implementation partner, works with manufacturers to pick the right first project, connect the systems you already run and put AI agents and tools into daily use, with people approving what goes out.
Assessments and consultations are provided by Genvo. How our partnership works.