← Back to overview

/ 01 - Discipline

Design.

Feasibility · optics · algorithms · mechanical & electrical

Before a single bracket is cut, our in-house engineers specify the camera, optics, lighting and algorithm around your part, your line speed and your tolerance. Designs are proven on sample parts in our lab and signed off as a buildable, testable specification.

Specify✦Feasibility✦Optics✦Lighting✦Sensors✦Algorithms✦Mechanical✦Electrical✦Bench-proven✦Specify✦Feasibility✦Optics✦Lighting✦Sensors✦Algorithms✦Mechanical✦Electrical✦Bench-proven✦

Across every
vision task.

How this discipline shows up on the line — inspection, measurement and code reading, handled by the same in-house team.

●A // Inspect

Designing imaging cells that catch the defect you care about — without tripping on the ones you don't.

  • Sample part study

    01

    Real parts imaged in our lab under candidate lighting, optics and algorithms to prove the defect can be separated from noise.

  • Lighting design

    02

    Bright field, dark field, dome, coaxial, backlight — selected and modelled around surface, finish and contamination type.

  • Optics & sensor spec

    03

    Lens, working distance, depth of field and sensor resolution chosen for the smallest defect at full line speed.

  • Algorithm architecture

    04

    Rule-based, blob, edge or deep-learning — selected for repeatability, false reject rate and audit traceability.

●B // Measure

Designing gauging stations with the repeatability and traceability your QA team will sign off on.

  • Tolerance & GR&R study

    01

    Measurement system analysis on representative parts so the gauge is proven before steel is cut.

  • Telecentric & 3D optics

    02

    Telecentric lenses, laser triangulation or structured light specified for sub-pixel and out-of-plane measurement.

  • Calibration strategy

    03

    Traceable calibration artefacts and on-cell calibration routines designed in from day one.

  • SPC / MES schema

    04

    Data model for every measurement, ready to drop into your existing SPC or MES without bespoke middleware.

●C // Read

Designing code reading and OCR cells that hit Grade A and hold it through peak.

  • Symbology study

    01

    1D, 2D and DPM samples graded against ISO standards to size lighting, optics and decode strategy.

  • Read rate budget

    02

    Per-station read rate targets calculated against line speed, parts-per-minute and downstream reject handling.

  • OCR / OCV strategy

    03

    Font, contrast and substrate analysis to choose between classical OCR and trained deep-learning models.

  • Track & trace data path

    04

    End-to-end data path designed from read point to serialisation database, including reconciliation and audit.

/ Next discipline

Build.