Welcome to Kunshan EEK Automation Equipment Co., Ltd.
Home/Materials /Denim AI Visual Inspection Equipment

Denim AI Visual Inspection Equipment

Denim AI visual inspection equipment targets indigo-dyed denim fabric, performing inline inspection of skew, holes, color differences, weaving defects, slub interference, cat's whisker indentations and other defects, with OK/NG output and production line interlocking.

Materials Overview

Indigo surface reflection + slub texture create double interference with judgement

Quick answers

Denim AI visual inspection equipment images indigo-dyed denim fabric, and the algorithm identifies defects such as weft skew, holes, color difference, weaving defects, slub interference, and cat's-whisker indentation, outputs OK/NG results, and interlocks with the production line. It is suitable for roll and cut-piece formats.

Denim is usually indigo-dyed with slub yarns (irregular thick and thin sections) and has a certain amount of surface reflection. The slubs themselves are normal texture but resemble weaving defects such as weft breaks and thick warp, so they are easily confused; and the reflection from the indigo surface swallows scratch and relief detail. These two points are the typical interference in denim inspection.

Defects specific to denim include weft skew (skewed fabric surface), holes, color difference (indigo batch variation), weaving defects (broken warp or weft, float stitches), abnormal slubs, cat-whisker indentation (finishing creases), and oil stains. Among these, weft skew is a geometric problem and requires angle measurement for judgement.

Denim is mostly used in garments and bags and luggage, with high output and fast material feed. For inspection, AI treats slubs as normal background and then identifies real defects; weft skew is measured using the edge and texture direction angle. Changing dye lots mainly requires adjusting the color datum.

Applicable Type

Raw Denim

Unwashed, the focus is on weaving defects and weft skew

Washed Denim

Whisker creases are present; process effects must be distinguished from defects

Stretch Denim

Contains spandex; positioning prevents stretching

Denim Cut Pieces

Already cut, inspected piece by piece with defect locating

Denim Roll Stock

Continuous material feed, line-scan full-width inspection

Common defect

Defect TypesTypical ManifestationsInspection Focus Points
Weft skewFabric texture direction is skewedAngle measurement judgement; a geometry issue
holeWarp/weft breakage forming holesBacklight Is the Most Direct
color difference Indigo batch variation, shade variationStable Color Reference
Weaving DefectsBroken warp/weft, skipped yarnDistinguish From Slubs
Slub AbnormalitySlub too thick or missingBuild a normal slub model
Whiskering / IndentationFinishing CreasesSeparate process marks from defects
oil stainOil Spots, StainsCompared against indigo background
Foreign fibersWhite Filaments, Off-color YarnHigh-contrast Features

inspection Workflow

Complete chain from loading to judgment

Visual inspection judgement chainThe complete chain from image acquisition to OK/NG judgement and PLC interlocking.image acquisitiontrigger captureTarget Positioningtemplate matchingfeature recognitionAlgorithm JudgementResult JudgementOK / NGindustrial communicationPLC interlockingRelease OKRejection / NG AlarmEvery step's result retains the image and judgement item, for traceability and review
Inspection Judgement Chain — Acquisition → Positioning → Recognition → Judgment → Communication interlocking; the entire chain runs locally on the machine.
  • 01 Denim loading / roll unwinding or cut-piece feeding
  • 02 Flattening and tension control
  • 03 Triggered imaging (suppresses glare)
  • 04 Texture orientation angle measurement (weft skew)
  • 05 Slub background modeling
  • 06 Defect segmentation and classification
  • 07 Defect Grading
  • 08 Synthesized OK / NG
  • 09 Marking / rejection and image archiving

AI Inspection Principles

AI Visual Inspection Principles

Denim inspection has to deal with both "reflection" and "slub interference". The algorithm first builds a model of the normal slub texture and a color baseline, isolating true defects such as missing weft, holes and color difference; weft skew is measured geometrically using the texture direction angle. If indigo reflection is not suppressed adequately, scratch-type defects cannot be imaged.

For illumination, diffused light provides even lighting and suppresses indigo reflection; low-angle light exposes cat's whisker creases and surface relief; backlight is used for holes. On washed denim the cat's whisker effect is a process feature, so the acceptance criteria must state clearly whether it is rejected.

For judgement, we recommend modeling the slubs as normal background and treating weft skew as a separate geometric item. The specific threshold and angle tolerance must be confirmed by measurement with OK / NG samples and the acceptance specification.

Applicable Industry

Applicable equipment

Common Question

How Do You Distinguish Slubs from Weft Breaks?
Slubs are the natural thick and thin places in yarn and are distributed regularly; a missing weft is a local absence of yarn with a break in form. A model of normal slubs must be built first, then deviations judged as weaving defects, using sample measurements to keep false calls low.
Can Weft Skew Be Measured Automatically?
Yes. The texture direction angle is used to measure the skew angle of the fabric surface, and anything beyond tolerance is judged abnormal. The angle tolerance follows the customer's acceptance criteria and requires sample measurement.
Are whisker creases on washed denim counted as defects?
Cat's whiskers are mostly a characteristic of the finishing process and are usually not rejected unless they exceed the design specification. Whether they are rejected should be made clear in the acceptance criteria to avoid treating a process feature as a defect.
Indigo Is Highly Reflective: Can Scratches Still Be Seen Clearly?
Dedicated illumination is needed to suppress indigo reflection (diffuse / low angle in combination), and low-angle light is then used to make scratches form a clear image. Whether reflection suppression is adequate is subject to measured sample testing.
What is the inspection accuracy?
To be added. Accuracy is determined by the field of view and camera resolution: pixel equivalent = FOV ÷ pixel count. First define the minimum defect size, then work back to the imaging solution and validate it by measurement with samples.
What samples are needed to develop a solution?
Provide OK and NG samples in different states such as raw / washed, the defect acceptance criteria, feeding format and production line cycle time.
Can It Be Integrated Into an Existing Production Line?
Yes. The system exchanges data with the PLC over I/O, TCP, RS485, Modbus, S7, Profinet and others, and outputs OK/NG to trigger rejection; the specific protocol depends on the site.

Submit sample testing

Send us OK and NG samples of denim in different states such as raw and washed, and we will run slub modeling and weft skew measurements and provide illumination plan and judgement recommendations.

JPG / PNG supported, multiple files allowed
Each file must not exceed 20 MB
    JPG / PNG supported, multiple files allowed
    Each file must not exceed 20 MB
      JPG / PNG supported, multiple files allowed
      Each file must not exceed 20 MB
        A solution engineer will contact you within 1 business day after submission

        Need to assess imaging conditions, defect criteria and cycle time item by item? Go to the Full Requirement Assessment →

        Send Us Your Material and We Will Show You the Measured Results

        Provide several OK and NG samples of the material, and a solution engineer will run actual imaging and judgement tests on the equipment to give a workable inspection configuration recommendation, rather than reading the datasheet alone.

        Online support WhatsApp
        WeChat
        Scan the QR code to add us on WeChat
        Call us Back to top