Selection Guides
Evaluate a machine vision camera supplier by application review quality, sample evidence, optical route support and integration handoff.
Direct answer
A machine vision camera supplier should be judged by how it handles real inspection evidence, not catalog size. Confirm six items: sample-image review, sensor route, interface support, lens match, lighting test path and integration handoff.
A useful machine vision camera supplier should ask for the part image, target feature, FOV, working distance, speed and acceptance criteria before quoting a camera model.
A large model list does not help if the supplier cannot discuss lens fit, light angle, trigger timing, integration output or sample evidence.
Share good, NG and borderline samples, current reference hardware, target tolerance, speed and expected support scope so the response can be evaluated for substance.
Before discussing resolution or brand equivalents, the supplier should ask what the station must prove, which samples are borderline and what output the line needs. Without that evidence, a camera recommendation is only a catalog guess.
Machine vision camera selection changes when the lens, light angle, exposure and fixture are reviewed together. A supplier that cannot discuss lens and lighting fit will struggle to solve reflective parts, small defects or measurement tasks.
Useful responses include assumptions, alternative routes, required sample evidence and integration notes. Weak responses only list part numbers without explaining how the camera will produce stable production evidence.
How to test before buying
A strong supplier response explains assumptions, missing inputs, alternate routes and sample-test evidence. A weak response only lists part numbers and leaves lens, light, trigger and support scope undefined.
Decision checks
Supplier asks for samples, FOV, speed, tolerance and failure mode.
Camera is reviewed with lens, lighting, trigger and I/O.
Supplier explains assumptions and sample tests before final model lock.
Decision table
| Factor | Practical rule | RFQ impact |
|---|---|---|
| Application review | Supplier asks for samples, FOV, speed, tolerance and failure mode. | Reduces wrong sensor and interface selection. |
| System route | Camera is reviewed with lens, lighting, trigger and I/O. | Prevents camera-only quotes that fail on the line. |
| Evidence quality | Supplier explains assumptions and sample tests before final model lock. | Improves buyer confidence and acceptance planning. |
| Export handoff | Supplier can provide RFQ notes, interface constraints and packing/delivery details. | Helps integrators avoid hidden commissioning gaps. |
Application proof
Small connector features require stable contrast, low distortion and repeatable part positioning before a pass/fail threshold can be trusted.
Battery tab weld inspection needs geometry confidence, stable position reference and traceability under high station throughput.
Reflective metal surfaces can hide scratches and burrs unless the lighting geometry is matched to the defect direction and part movement.
Common mistakes
Factory handoff
The factory route review starts by checking whether the image can be made stable with lighting and fixture control. Then the camera, lens, reader or 3D sensor route is sized against speed, resolution, interface and installation constraints.
If you already have a Keyence, Cognex, Basler, OPT, LMI, Hikrobot or barcode-reader reference, include it as a reference model. Deyi Vision uses it to understand the application class; final selection still depends on real samples and production limits.
Guide to RFQ
Use the guide to frame the question, then send the details so engineering can recommend a route.
Request engineering RFQGuide FAQ
Ask how they will verify FOV, smallest feature, working distance, interface, lens, lighting and line speed before recommending a camera model.
No. For production inspection, application review quality and sample evidence matter more than catalog size.
Send part photos, good and bad samples, target tolerance, line speed, working distance, FOV, current model reference and output requirements.
