Machine vision basics: visual inspection and optical quality control
Machine vision basics are often mixed up with visual inspection and optical quality control. These three terms describe adjacent but distinct layers of the same system: the technology, the task, and the process. This article sorts them out, shows where they overlap and where they do not, and leaves you with a clear starting map for your first machine vision project.
It is written for production and quality leaders who need to run a machine vision workshop today, without first spending three weeks on technical literature.
What machine vision actually means
Machine vision is the umbrella term for systems that combine cameras, lighting, compute and software to automate visual decisions on a production line. The decisions can be pass or fail (quality), which part is this (identification), or where exactly is the part (robot or gripper guidance).
Three components are always in play: an image source (camera plus lighting plus optics), an algorithm that interprets the image (rule-based image processing or a neural network), and an action the plant floor takes on the result (reject, document, stop the line, trigger a gripper). For the full picture of how these pieces fit together we recommend the industrial image processing guide and the machine vision inspection guide as companion reads.
Visual inspection: the applied use case
Visual inspection is the classic task most people mean when they talk about quality with machine vision. When a camera on a packaging line checks every label, that is visual inspection. When a camera inside an injection molding machine flags flash or inclusions, that is visual inspection. The term describes the task, not the technology.
In practice visual inspection is the bridge term quality managers understand immediately. They have done visual checks manually or with magnifiers for decades. A camera that automates the same check is an inspection station. For a deeper definition read What is AI visual inspection.
Optical quality control: the process term
Optical quality control is the quality-management process term and describes the procedure, not the tool. It covers planning (what is inspected, with what tolerance), execution (manual, optoelectronic or AI-based) and documentation (who accepted what, when, against which spec).
The important distinction: optical quality control can exist without machine vision. A trained operator with a magnifying lamp is doing optical quality control. What machine vision adds is repeatability, documentation per part, and the ability to run 24/7 without drift. If the goal is audit-grade traceability under ISO 9001 or a GMP regime, it is the optical quality control process that matters, and machine vision is one of several execution options.
Four process classes to know
Every machine vision project falls into one of four algorithmic classes. Knowing which one fits changes the data you need to collect and the lighting setup you will build.
Rule-based image processing. Classic measurement, edge detection, OCR and blob analysis. Deterministic, fast, documented for decades. Still the right choice for dimensional measurement, presence/absence and code reading when geometry and lighting are stable.
Deep learning classification. A neural network sorts each part into one of several classes (pass, scratch, dent). Works well when the defect is visually consistent and examples exist for every class. This is where modern AI visual inspection lighting practice pays off most, because contrast drives model accuracy.
Anomaly detection. The model learns what good parts look like, then flags anything that does not match, even defect types nobody has seen before. The right pick when defects are rare, varied or non-repeating.
Object detection and segmentation. The model locates where each feature sits inside the image, pixel by pixel. Used for counting parts on a tray, isolating individual cells inside a batch, or guiding a robot to pick a specific item.
Four factors that decide a first project
Lighting. The single biggest predictor of whether a machine vision project works. Lighting makes a defect visible or invisible, and no algorithm recovers contrast that never existed in the image. Budget for a week of lighting trials before you touch the model.
Data. Classical algorithms need carefully chosen parameters. Modern AI needs examples, usually 50 to 500 labelled images for a pilot. The right number depends on defect variety; more classes means more images.
Integration. A machine vision station that cannot talk to the PLC, MES or the reject flap is expensive decoration. Plan early how inspection results reach the line, including retries, fault states and the operator HMI.
Maintenance. Every line changes over time. Models drift, lighting ages, cameras get dust. A working station today is not a working station in six months unless someone owns it. Budget 10 to 20% of build cost per year for ongoing care.
Three applications that ship fastest
Label and print inspection on packaging. High volume, clear rules, lots of existing reference imagery. Often the first line where the ROI proves itself within weeks.
Seal and fill checks on bottles, pouches and blisters. See our deep dives into food and beverage packaging and pharma packaging for category-specific playbooks.
Surface defects on injection molded and stamped parts. Well understood defect classes, stable geometry, easy to light. A good first project for SMEs that do not want to start with the hardest problem on the floor.
Where to start
Pick a line that runs every day, with a defect class your operators can describe in one sentence. Build a small lighting rig and capture 200 images. Decide between a rule-based and a learned approach only after you have looked at your own images. For a broader technology-by-technology overview also read the machine vision systems guide.
If you want to compare notes with other manufacturers working on their first or fifth project, join the Enao community at enaovision.com/#community. You will find people who have shipped the defect class you are about to tackle, and who are usually happy to save you a week of trial and error.