Solve a machine vision engineering task

Start with the task

The library is organised by engineering decision, not by product. Choose a direction, note what cannot change on your line, and keep assumptions apart from measured and datasheet values.

What you will find

• Selection guides that state their limits.
• Calculation methods that use the same inputs as the MVisionPro tools.
• Step-by-step checks for faults and unstable results.
• Comparisons and cases that name their evidence.

Directions

• Plan, accept and maintain a vision project — Write down the inputs, the acceptance criteria and repeatable checks before anyone calls the equipment or the software finished.
• Choose optics and control blur and distortion — Go from the field of view, the smallest detail, the distance, the motion and the light to a camera and lens you can verify.
• Get images from the camera to the computer without losing frames — Check link capacity, cable limits and the connection layout first, then the computer: its processing, storage and response time.
• Make codes read reliably on a real line — Find out why barcodes and Data Matrix codes read unreliably: the image, the marking, the pixels per code module, the reader settings or the verification conditions.

Latest guides

• Choosing machine vision lighting for DataMatrix codes marked on metal — Try low-angle dark field on recessed dot-peen marks, a diffuse dome on curved or mirror-like metal and coaxial light on a flat reflective face, add crossed polarizers where glare remains, and keep the setup that reads your worst parts across their poses. A clean decode is not a grade: where your application specification requires one, a verifier grades the mark to ISO/IEC 29158:2025 or to the edition the specification names.
• Deciding whether a measurement needs a telecentric lens, and choosing one — Estimate how much a regular lens changes the measured size when a part sits off the calibration plane, and plan a telecentric lens when that estimate alone uses up your measurement budget or an angled view hides an edge that a straight-on view would show. Choose it by magnification, since the field is the sensor size divided by magnification, check field, depth and image circle against the part, and accept it on the calibrated error measured at both height extremes.
• How to calculate lens focal length and field of view when the working distance is fixed — Calculate the focal length from the active sensor size, the field you need and the working distance counted from the lens principal plane. Then check the real field and pixels per millimetre of the catalog lens you pick.
• Replacing a fixed Cognex DataMan or KEYENCE SR reader: what has to match — Record what the installed reader delivers, from codes and field at the working distance to PLC data, timing, wiring, power and mounting, and check each item against one exact variant of the new reader. Our analog tool screens camera parameters only, so treat the models it shows as starting points and decide the swap with a paired test on your parts and PLC.
• Setting up a camera to inspect bottle caps, fill level and labels on a conveyor — Find a view and light that show each cap, fill-level and label defect, starting with a side backlight for cap height, tilt and a visible liquid line and with front or dome light for cap color and print; checks can share a frame that shows everything they need, and a separate view is for a hidden side, clashing lights or too little field or resolution. Derive belt speed from bottle rate and pitch and the exposure limit from sampling and blur allowance, pulse the light where that limit falls below a camera's minimum exposure, and choose between a PC camera and a smart camera on software, inputs and outputs and timing measured on the line.
• Checking whether a lens fits your camera: image circle, sensor diagonal and mount — A format label such as 2/3″ names a size class, not a measurement. Compare the lens image circle with the sensor's active diagonal in millimetres, then check mount, thread and flange distance separately.
• Choosing a camera interface when several pass the bandwidth check — Several interfaces usually pass the bandwidth check. Choose between them by cable run, one link per camera or a shared network, power on the cable, and the hardware in the PC.
• Choosing an industrial code reader: fixed or handheld, label or direct part mark — Decide four things: what moves, label or direct mark, code type with module width, and the link to the controller. Then prove the exact variant with a read test on real codes.
• Deciding how many pixels a feature needs before you choose a camera — Spotting a feature, measuring it and rejecting parts on it each need their own pixel budget. The calculator gives the sampling, while detection, measurement error and reject rates come from tests on real parts.
• Deciding whether a rolling shutter is good enough for your moving part — A moving part does not rule out a rolling shutter. Shift the reference points by the part's travel between row times, recompute the dimension and budget blur separately.
• Deciding whether depth of field covers every part position — Pick the lowest f-number whose calculated depth of field covers every part position plus your reserve. Then prove it with the real decode or measurement at the nearest and farthest positions.
• Line-scan image stretched or compressed: how to find the cause — Image a known-distance target and measure both axes before changing any setting. The difference shows stretch or compression, and a check of each axis tells whether triggering, the encoder or the optics changed.
• Setting exposure for a moving part, and deciding whether to pulse the light — Turn line speed and the blur you accept into a time limit. Then decide whether it limits the camera exposure or a synchronized light pulse, worked through for a code on a conveyor.
• Why a code does not read, and what to check first — Save a failed frame and a good one, then change one thing per test. A smear suggests motion blur, while a sharp, well-framed image that fails points to the decoder.
• Working out the camera resolution a code needs — Divide the code's size, without its quiet zone, by its module count to get the module size. Size the camera from your decoder's pixels-per-module limits, then recheck them with the installed lens.

Related

How this material was prepared

Prepared by MVisionPro from the stated sources and the MVisionPro calculation engine. A physical test is claimed only when the material says so explicitly. Read the editorial method.

Editorial status: verified. Content updated 2026-09-23.