Configuring LumiScan Area Detector for Region Inspection

When LumiScan marks too few or too many image regions, changing size filters first usually hides the symptom instead of correcting the brightness-based separation. The Area Detector identifies regions by their brightness difference from the background, then evaluates region properties; correct the image separation before tuning filters.

Stop changing size filters before checking region separation

Two quick adjustments often make an inspection look better on one image while making it less reliable across normal part and lighting variation:

• Loosening area or dimension filters: This can admit small noise regions, but it does not correct a target that blends into the background or a background region that has been detected as a target.
• Tightening filters to remove extra regions: This may suppress an unwanted region in the current image while also rejecting a valid small, incomplete, or differently positioned part.
• Changing several detection criteria together: If the overlay improves or worsens, you will not know which change affected the result. Keep a known-good image available and adjust one criterion at a time.

Check the live image and detected-region overlay first. If the target and background are not separated into the intended regions, revisit brightness-based detection settings before filtering by area, width, height, position, roundness, or brightness.

Read the live overlay to locate the failure

LumiScan Area Detector separates image regions from the background using brightness differences. It then evaluates each detected region by properties including area, width, height, position, roundness, and brightness. This creates two distinct questions: did the image produce the right regions, and did the configured criteria accept the right regions?

Use the overlay and displayed measurement data in LumiScan Cockpit to distinguish detection from evaluation. Do not infer an exact threshold, unit, or pass/fail rule from the overlay alone; read the actual configured setting and measurement shown for the inspection.

Set up a representative image before tuning

Choose an image that includes the expected part or material and the background it must be separated from. For a completeness or defect check, use examples that show the relevant condition, such as a missing part, hole, contamination, excess adhesive, or surface damage, when those conditions are part of the application.

• Confirm that the target area and background are visible in the live image.
• Use consistent, representative imaging conditions while comparing adjustments. If brightness or image appearance changes, recheck separation before reusing filter values.
• Keep examples of expected detections and known unacceptable conditions available for comparison. A setting that works on a single image may not separate normal variation from a defect.
• Record the relevant setting and observed overlay after each meaningful change so you can restore a known state.

The source of an image change may be illumination, material appearance, object position, or another part of the imaging setup. The app description does not prescribe a lighting arrangement or numerical image-quality limit, so use the live image to identify whether the target-background brightness difference remains usable.

Adjust brightness criteria before region filters

Work in LumiScan Cockpit, the configuration and monitoring interface for the native ctrlX app. Begin with brightness-based region separation because that step determines which areas become candidate regions.

1. Display the live image and detected-region overlay.
2. Adjust the brightness threshold or relevant detection criteria until the intended target areas appear as regions and the background is not being treated as a target.
3. Change one brightness-related setting at a time. Recheck the target and background after each adjustment.
4. Repeat the check on representative images, including expected variation in parts or materials. Do not settle on a setting that only works on one favorable image.

The available application description identifies adjustable brightness thresholds and other detection criteria, but not threshold polarity, a numeric range, or a particular parameter name. Read the controls displayed by your installed app and verify their effect in the overlay rather than assuming that a higher value always admits more or fewer regions.

Filter detected regions against the inspection objective

Once the overlay shows the right candidate regions, apply filters to distinguish the regions that matter. The Area Detector can evaluate area, width, height, position, roundness, and brightness. These properties support presence checks, completeness verification, material identification, position checks, and defect detection, but the correct combination depends on the object and inspection requirement.

• Presence or completeness: Check that expected part regions are present. Use observed region dimensions and position to distinguish a complete target from a missing or incomplete one.
• Position: Compare the detected location with the required placement. Set acceptance criteria from the machine’s inspection requirement, not from an arbitrary coordinate margin.
• Shape or roundness: Use shape-related measurements when they distinguish the target from an unwanted region or when shape is itself the inspection feature.
• Defects: Look for unexpected regions associated with the defect being inspected, such as contamination, holes, missing material, excess adhesive, or surface scratches. Confirm the intended defect appears as a candidate region before tuning its filters.

Change one filter at a time and watch which regions disappear or remain. A filter is useful only if it separates valid from invalid cases without discarding acceptable product variation. If a valid and invalid region overlap in the measured properties you are using, revisit the image separation or the inspection method rather than continually tightening the same filter.

Validate the decision on both accepted and rejected cases

Live visualization is a setup aid and diagnostic view; it does not by itself prove that the inspection makes the required decision reliably. Verify the full sequence: the target becomes a detected region, its measurements are evaluated as intended, and the inspection result matches the known case.

1. Check an acceptable example and confirm that required parts or features are detected and accepted.
2. Check a representative unacceptable example, such as a missing or incomplete part or a relevant surface defect, and confirm that the region or missing-region condition produces the intended rejection.
3. Repeat with representative material and image variation. Review the live overlay and measurement values whenever a result differs from expectation.
4. Recheck after changing brightness criteria or filters. A change that fixes one case can alter other region detections.

Keep a record of the images used, settings changed, measured properties, and observed result. This makes it possible to tell whether a later failure comes from changed image separation, filtering, or the inspection requirement itself. Do not claim a production acceptance rate from a few setup images; evaluate the range of actual operating conditions required by the machine.

Use ctrlX integration without assuming undocumented controls

LumiScan Area Detector is described as a native app for Bosch Rexroth ctrlX OS and part of the LumiScan Vision Suite. Configuration and monitoring take place through LumiScan Cockpit. Use that interface for the available detection settings and live diagnostics; do not assume an undocumented control, communication tag, output address, or trigger sequence.

The app provides region overlays and measurement data, but the provided product information does not specify PLC interface details, result data formats, timing, camera configuration, or particular software versions. If the machine depends on those details, read the installed app documentation and the actual ctrlX project configuration before connecting inspection results to machine logic. Test the complete signal path using the application’s documented interface.

FAQ: Configure LumiScan Area Detector