Application of Artificial Intelligence in Textile Defect Detection

 

AI made textile defect detection more accurate and consistent than old inspection methods. AI machines quickly spot fabric problems. They detect holes, stains, weave issues, strange patterns, and dyeing faults. They use computer vision and machine learning algorithms for this task. Artificial intelligence is helping the textile sector. It inspects textile defects and offers solutions. Today we will share Artificial Intelligence in Textile Defect Detection.

Introduction to Textile Defect Detection

The textile industry has multiple stages for preparing fabric, that is, to wear. We picked cotton from the fields. Then, we sent it to spinning units to make yarn. After that, it went to weaving to create fabric.

We take woven fabric to the processing industries. There, it goes through various chemical processes. These processes clean, dye, and print the fabric. It also includes the mechanical effects of calendering and sanforizing. These processes help achieve the desired finish on the fabric.

Fabric preparation is a long journey with many steps. During these steps, the fabric can face failures and defects. We address and fix these issues during processing.

Let’s see the major defects detected that may occur during the processing.

 

Yarn Defects

Broken Filament

It occurs when the individual filaments constituting the main yarn are broken.

Colored Flecks

It is the presence of colored foreign matter in the yarn, called colored flecks.

Knots

It occurs when the individual filaments constituting the main yarn are broken.

Slub

A slub is a thick, bumpy spot in the yarn. Here, the fibers twist loosely or not at all. This makes it wider than normal yarn.

Weaving Defects

Broken Ends Woven in a Bunch

This defect happens when broken or loose threads get woven into the fabric. This creates an uneven or flawed area.

Broken Pattern 

A broken pattern occurs when the weave or design of the fabric interrupts. This makes the look uneven or incomplete.

Double End

This defect appears as a thick line running along the length of the fabric (parallel to the warp). It occurs when two or more yarns become intertwined as if they were a single thread.

Float 

A float occurs when the warp and weft threads in the fabric do not interlace correctly. This causes the threads to sit loosely or incorrectly on the surface.

Gout

Gout means small bits of foreign material, like lint or debris. These bits become woven into the fabric during manufacturing due to an oversight.

Hole, Cut or Tear

Generally, the fabric shows a cut or a hole. It may be due to many factors.

Lashing In

A piece of extra yarn gets woven into the fabric near the edge (selvedge) by mistake during production.

Local Distortion

Distortion happens when the warp (vertical) or weft (horizontal) threads in the fabric shift from their usual straight lines. This can cause the fabric to look uneven or misaligned.

Missing Ends

The fabric has a gap that runs parallel to the warp (vertical threads). This occurs when one or more warp threads are missing in that area.

Oil or Other Stains

These are small defects in the fabric. They happen from stains like oil, rust, grease, or other substances.

Oily Weft

Dirty or oily streaks run across the width of the fabric, either covering the entire width or part of it. Contaminated weft (horizontal) threads during weaving cause these marks.

Oily or Soiled Ends

These are oily or soiled warp threads.

 Reed Marks

A gap or crack in the warp (vertical) direction comes from a damaged reed. The reed is a tool in weaving that maintains even spacing between threads.

 Selvedge Defect 

These are different defects that appear at the selvedge.

Slough off

A slough-off occurs when weft (horizontal) yarn tangles or bunches up. This yarn then becomes woven into the fabric due to an error.

 Smash

Ruptured cloth structure characterized by many broken warp ends and floating picks.

Stitches

A single thread floats either in the warp or weft. It is very prominent in the case of different colors of the warp and weft.

Snarls

A short piece of yarn, often the weft, twists or loops back on itself. This creates a small knot or snarl in the fabric.

Untrimmed loose threads

We term any hanging threads in the fabric as loose threads.

Weft Bar

An unwanted bar is a visible line that runs across the width of the fabric. It looks different from the normal fabric around it. This defect often stands out due to its distinct appearance.

 

Processing Defects

Blurred or Dark Patch

A blotch on printed fabric causes a smudged patch. This makes the design look unclear or uneven.

Bowing

Bowing is a condition of the fabric where the warp and weft yarns do not remain at right angles to each other.

Dye Bar

A bar is the weft direction due to the difference in color or shade of dye used.

Dyestuff Stain

An unwanted color mark on the fabric qualifies as a dye stain.

Hanging Threads

A break in the fabric pattern happens when hanging threads mess up the design. This creates a gap or misalignment in the print.

Miss Print

A misprint can cause one or more of the following:

• The printing is not as per the required design.

• The outlines and colors in the design do not remain in their proper places.

• A bare place without any printing.

Patchy, streaky, or uneven dyeing

The fabric has uneven dyeing, with areas that appear lighter or darker along or across its width. This can appear as patches, streaks, or lines of different shades. It may also include differences in dye intensity on the edges (selvages) of the fabric.

Shading or Listing

This refers to a gradual change in color shade across the fabric, which can occur in two ways. The first point centers on the selvedge. Here, the shade changes gradually from the fabric’s middle to one edge (selvedge).

The second is selvage to selvage, where the shade gradually shifts from one edge to the opposite edge. This leads to uneven or inconsistent color on the fabric. Some areas may look different.

Uneven printing or tinting.

In a printed fabric, the design looks bold and clear in some spots. Yet, in other areas, it appears faint, blurry, or unclear. This inconsistency in the print creates an uneven appearance in the fabric.

Water Mark

A watermark refers to an unwanted ripple effect (light mark) produced on the fabric.

Bleaching Spot

The fabric is characterized by a yellow tint in the bleached cloth. In bleached goods, such a difference is not acceptable.

Mechanical Defects

 

Pilling

Washing, dry cleaning, or wearing can tangle fibers. This leads to pills or balls forming on the fabric’s surface. These areas are dense enough for light to pass through.

Uneven or Loose Piles

Uneven or loose piles happen when the height of the pile (the raised surface) varies across the fabric.

In fabrics with a raised texture, inconsistent raising can cause uneven patches. This creates a patchy or irregular appearance on the surface.

Piles Less Spot

In terry and velvet fabrics, a pile-less spot is where the soft, raised surface is missing, even if it should be there.

Uneven raising of fabrics can lead to a patchy look. This makes some areas stand out from the rest. This results in an inconsistent texture or finish.

Uneven Milling 

Uneven matting together of fibers results in uneven milling.

Mill Rigs

Milling produces creases known as mill rigs.

Importance of Defect Detection in Textiles

Detecting fabric defects in textile industries is crucial. A small mistake can cause big losses and harm the fabric’s quality.

Timely defect detection saves time, energy, and resources in fabric processing. It also boosts the quality of the finished cloth.

For example, we have learned that digital printing requires careful handling. changing the padding thickness caused us severe shade-change issues.

In textiles, lab staff and technicians monitor each step with great attention. Then, they move on.

Traditional Defect Detection Methods and Limitations

The textile industry performs the fabric defect detection process. The traditional methods used by labor are as follows:

1. Manual Handling

2. Lenses Inspection

3. Light Board

4. Mechanical Sensors

5. Physical Methods

Power loom fabric has various qualities. We need to check these for quality assurance. We check these fabrics before processing, and we need to solve defects upon finding them.

Manual Handing

The team checks the fabric and power looms from the weaving department by hand. The team in the mending department then takes the greige fabric to mend it.

Someone places the entire length of fabric on a tilted light board. This helps to spot hanging threads, stains, and other defects. Someone manually cuts the long threads using a cutter.

Lenses Inspection

Lens inspection checks the fabric’s construction. Two types of lenses commonly found in textile units are 1 inch and 1/2 inch. Both lenses verify the fabric quality and construction.

The half-inch lens saves time. You count half the threads, which makes it easier to check different quality types.

Using a lens, a technician can count the number of threads in the warp and weft sides. This is the most traditional method for evaluating the quality of greige fabric.

Light Board

The tilted light boards are used to determine stain defects, hanging threads, and fabric width. In the dyeing department, a light box is also used to check and match the shade of the fabric under different light sources, such as TL84, F11, and UV light.

Mechanical Sensors

Mechanical sensors monitor fabric quality on various machines during processing. The skew bow machines use sensors that remove the skew bow of the fabric. Mechanical sensors also detect uneven fabric thickness and texture issues.

Physical Methods

Physical testing methods are old, traditional methods for finding defects during fabric processing. For example, after desizing the fabric, we check the size percentage to see if it has been removed.

After de-sizing, the fabric goes to the machine, where I check its degree of mercerization. Then, the staff monitors the operation in the bleaching department.

After bleaching, the bleached fabric should penetrate more than 3 cm. If it does, it’s good for printing or dyeing. After printing, we take a fabric sample from the soaper machine. This helps us check for color bleeding.

After finishing the fabric, it moves to packing. A sample is taken for soaking and washing. This checks if the color bleeds.

 Limitations

The traditional methods for defect detection have limitations. Workers use these methods manually, which may cause human error. The manual methods are time-consuming and inconsistent due to fatigue.

Lens inspection is also time-consuming because workers count the number of threads individually. This method is not suitable for large quantities.

AI-Based Techniques for Textile Defect Detection

The different fabric structures result in various textures and other types of defects. The combinations are many. This makes automated fabric detection a real challenge for the textile industry.

Manufacturers have workers check fabric by eye. But this method can be inconsistent and unreliable.

Artificial technology created the Wise Eye. This smart defect detection system works on any weaving machine.

It helps manufacturers detect and correct travel defects during the photo process. Let me say that with a wise eye, fabric defects have nowhere to hide.

Imagine the fabric’s complex structure as an image’s regular background texture. Defects are foreground objects. They disrupt the normal texture of an image.

What we do is single out abnormalities from the complex background texture. The system features a powerful LED light bar and a high-resolution CCD camera.

They are mounted on a rail and driven back and forth along it by an electronic motor. During the weaving process, the camera captures images of the whole width of the woven fabric.

We process the images first. Then, they go into an AI machine that uses a vision algorithm to find fabric defects. When a defect shows up, the smart eye quickly alerts the weaving machine or the factory’s info system.

Wise Eye can detect 40 common fabric defects with a detection rate of over 90%. Wise Eye successfully integrates AI, Big Data, and deep learning technologies in a new way.

This tech breakthrough helps the textile industry. It also marks a big step in automating quality control.

Implementation and Benefits of AI in the textile industry

The fabric inspection process depends on people. This can waste time and cause mistakes. Many industries use the AI virtual inspection system. It works around the clock without needing a person.

You can get rid of manual fabric inspection work. It runs 24 hours a day without any workers. It has saved the factory from a lot of manual labor. Additionally, you can use it on many types of single-color fabrics.

Advanced machine learning technology finds many defects. These include broken ends, broken filling, knots, holes, dirt, fuzziness, and weft shrinkage.

AI-powered machines catch defects at a range of 0.1 mm with over 85% accuracy.

                                         AI System Catching Fabric Defects

The AI system can learn and improve its defect detection. This leads to better flaw identification by connecting to your ERP system.

Get the inspection report and data-driven results in real time. This will provide a significant increase in your efficiency.

Future Trends and Innovation

In 2025, technology is undergoing rapid changes. We’re about to see a revolution where tech fits into our daily lives.

Technology powered by artificial intelligence and smart fabrics is leading this exciting transformation. These advancements will improve our health, fitness, and daily lives like never before.

Picture clothing that monitors your heart rate, tracks your workouts, and adapts to your body temperature. This is not just science fiction anymore.

We are rapidly approaching a new era. By weaving technology into our daily lives, we connect our bodies to the digital world.

Let’s dive into the exciting world of smart fabrics and AI. They are set to change how we live.

Smart Fabrics are revolutionizing the textile industry. These fabrics are more than just fashion. They are smart, responsive, and can interact with us and our surroundings.

                                                         AI-Powered Textiles

More Applications of AI

Picture workout clothes that track your heart rate and muscle activity as you exercise. Also, think about clothing that adjusts to the weather, ensuring your comfort.

Companies like Google, with its Jacquard project, are leading this revolution. Nike uses sensors to check athletic performance. Under Armour makes clothes that regulate temperature.

Hexskin makes biometric shirts that capture physiological data. Here are a few companies leading the way in smart fabric innovation.

Smart Fabrics provide the platform; artificial intelligence drives their transformative potential. AI algorithms analyze data from sensors, offering insights into our health and well-being. This real-time analysis provides personalized feedback, empowering informed health decisions.

AI excels at identifying patterns and making predictions, revolutionizing healthcare. It helps spot health problems early, which is key for chronic conditions. Fitness AI makes tailored training plans and boosts performance.

AI also monitors mental health by detecting stress, anxiety, or depression patterns. Smart fabrics and AI will transform our daily lives in profound ways.

Imagine a future where we manage chronic diseases early. This will stop complications and boost the quality of life. The potential benefits of this technology extend beyond individual health.

Smart Fabrics collect data that we can combine and anonymize. This offers useful insights for medical research and public health efforts.

                                                             Textile Shrinkage Tester developed based on Artificial Intelligence and Visual Inspection Technology

 Textile Hydrostatic Pressure Tester with Artificial Intelligence Technology

Conclusion

Artificial intelligence is changing the textile industry. It helps detect defects in fabric. This technology reduces the need for manual inspections.

It worked 24/7 with no workers needed. It saved the factory from a lot of manual labor and can be used on many types of single-color fabrics.

The advanced machine learning technology spots many defects. These include broken ends, filling knots, holes, dirt, fuzziness, weft shrinkage, and more.

For more information on textile testing methods/standards
or textile testing machines, contact us:
What’s App: +86 180 2511 4082
Tel: +86 769 2329 4842
Fax: +86 769 2329 4860
Email: [email protected]