Glossary · Machine vision
Machine vision
Also known as: Industrial image processing, Industrial vision
German: Industrielle Bildverarbeitung
In industrial automation, machine vision is the use of cameras, lighting, optics and image processing to automatically inspect, measure, identify or guide in manufacturing and logistics, with results passed to the machine or plant control.
- Machine vision
In one sentence
Machine vision uses cameras, lighting and image processing to inspect, measure, identify or guide automatically in production.
Example
Machine vision on the assembly line checks connector positions, reads serial numbers and guides the robot to each screw hole.
How it applies
- Engineering: A machine vision project defines the task, then designs lighting, optics and camera to make the relevant features visible, and only then the software. Typical tasks are inspection, measurement, identification (codes, text) and robot guidance.
- Operation: Machine vision results often decide over reject, rework or robot motion, so the system's reaction to errors, missing images and timeouts must be defined.
- Documentation: The documentation team should document the vision tasks, validation results, settings and change procedures, and provide operators with guidance on cleaning, reference parts and typical faults.
Machine vision vs. computer vision
Computer vision is the broad scientific field of making computers interpret images. Machine vision is its industrial application with defined lighting, fixed cameras and real-time integration into machines. Many AI methods, including Neural network models, are shared.