A machine vision system used on a production line may need to do more than identify a single defect. It may also need to inspect different products, read codes, measure dimensions, guide a robot or preserve images for troubleshooting. The XG-X Series is designed as a configurable vision platform for these varied requirements, combining multiple camera and lighting options with inspection software, controller tools and data-management features.
For automation engineers and integrators, the practical consideration is how to match imaging and processing methods to the inspection task while keeping setup, changeover and maintenance manageable. A platform that supports different approaches can help address current applications and accommodate changing production needs without treating every inspection as a separate system.
Match the imaging method to the inspection challenge
Different parts and defects require different image information. The platform supports area cameras, high-resolution cameras, line scan cameras and 3D profilers or cameras. This gives system designers options for inspecting small features, large surfaces, continuously moving material and height variations.
High-resolution cameras can capture detailed images across a wider field of view than lower-resolution alternatives. The source specifies cameras up to 21 megapixels, with images used for tasks such as checking printed features on IC trays. Line scan cameras build an image line by line, which can be useful for large or continuous targets. The described line scan options include resolutions up to 8K; the whitepaper also notes a 67-megapixel image configuration for a line scan application.
For features that are difficult to distinguish by color or grayscale alone, 3D inspection adds height information. Combining 2D image intensity with 3D height data can help identify conditions such as lifted screws, chipped plastic or variation in a solder fillet. Height measurements and relative-height tolerances can then be used to assess whether parts meet inspection criteria.
Machine vision system lighting and image processing
Lighting can determine whether a feature is visible consistently. The LumiTrax™ approach captures images under controlled illumination from multiple directions and analyzes pixel-brightness changes to separate surface shape from texture. This can help reduce the influence of patterns, glare or surface finish when inspecting features such as embossed characters, package perforations and metal-surface defects.
Multi-spectrum imaging provides another option for applications where slight color differences matter. The described approach combines a high-speed monochrome camera with lighting at multiple wavelengths, allowing image capture under different lighting colors. Separate images can support different inspection needs—for example, viewing a printed pattern for position correction while using another illumination color to assess a defect.
For line scan applications, the LumiTrax specular reflection mode can generate different image types from captured data, including specular reflection, diffuse reflection, gloss-ratio and shape images. Selecting the image most suited to a defect can help distinguish conditions such as dirt, linear flaws, dullness or dents. The key design point is that imaging mode and lighting should be selected around the material, surface and target defect rather than relying on one image-processing method for every part.
Build and maintain inspection logic with flexible software
VisionEditor uses flowchart-based programming to organize inspection steps. Users can build flows by arranging processing instructions, view captured images and processing results, and use simulation and debugging functions during development. Variables can be defined for different scopes and types, with comments to support program review. The software also provides calculation and script functions, including error checking to help locate syntax issues.
The available inspection tools cover tasks such as pattern search, edge measurement, defect detection, blob analysis, color evaluation, OCR and 1D or 2D code reading. Some tools analyze grayscale information rather than relying only on binary thresholds, while profile-based tools can identify edge deviations such as burrs or chips. Filtering options help narrow detected defects by characteristics such as size, shape or intensity.
These tools can be combined in one inspection flow. For example, a program might locate a pattern, measure an edge and check a code, with results available for the machine’s control sequence. The appropriate tool combination depends on the application and must be validated against actual parts and operating conditions.
Support product changeover and troubleshooting
Production environments often require inspection settings to change between products. The system supports background setting changes, with the whitepaper stating a 5 ms program-setting changeover under specified conditions. It also cautions that restrictions apply and that frequent changes may affect SD card service life, so users should consult the manufacturer’s guidance and back up settings regularly.
An online retest function allows archived images to be used to evaluate program adjustments while inspection processing continues. This can help teams investigate a failed judgment and check revised settings without repeatedly inspecting parts. It is a troubleshooting aid, not a substitute for validating changes against production requirements.
Remote connectivity adds another maintenance option. VisionTerminal can monitor connected controllers, log measurement results, transfer image and setting files, and provide access to trace information about processing units, I/O signals and commands. Remote access can support diagnosis without requiring an immediate site visit, subject to the plant’s network, access and operational requirements.
Connect inspection to robots and production data
For vision-guided robotics, the platform provides communication options for robots from multiple manufacturers and supports functions such as jog operation and automatic calibration. Calibration can reduce reliance on manual configuration, but the complete system still depends on correct camera, robot and workcell setup.
Optional VisionDatabase software supports organizing inspection images and results, linking records to external identifiers such as serial or lot numbers, and searching stored data. It can also support comments, graphing, export and settings backup. These capabilities can help teams review inspection history and use stored images during troubleshooting or program evaluation.
Conclusion: design around the inspection and the operation
A configurable machine vision system can bring 2D and 3D imaging, specialized lighting, inspection tools, robot integration and diagnostic functions into a common platform. For plant teams, the value lies in selecting the right combination for each task and planning how programs will be changed, validated and maintained. Start with the defect or measurement requirement, confirm that the imaging method makes the relevant feature visible, and then build an inspection and support workflow that fits the production process.