The textile industry is considered to be one of the most competitive industries in the world. This is solidified by the fact that the global textile exports were estimated at 238 billion USD in 2016, according to WTO. In 2017, the Asian Pacific region was the largest to manufacture textile. It accounted for 68 percent of the total market share.
The competitiveness of this specific industry truly makes quality control important. In the industry, we see buyers judge their manufacturers on the ability to deliver high-quality textiles at low prices. Efficiency is at the core.
Thus, manufactures need to institute processes in quality control from start to finish in production. This is to make sure that final products are actually of high quality.
Manufacturers need to make sure that raw materials are rightly sourced. They also need to check the perfection of fabric construction. What is more, they also need to see that the finished material comes with almost no defects. Such considerations are important for companies that seek to maintain their share of competency in the industry.
Textile includes different materials made out of fibers. These include hemp, cotton, silk, wool in natural terms. The synthetic fiber material includes rayon, spandex, nylon, and polyester, for example. It is important to inspect textiles during production. This is to make sure that the finished goods that are delivered do not have any defects in the surface or structure. This would pretty much save the expenditure by 45 to 65 percent.
However, it is important that the production procedures for textiles vary based on end-use. It might include insulation, apparel, home decoration, and automotive interiors. The risks for the defects that come through a given textile item are important. For instance, invisible or visible defects might get introduced for a given textile product.
For instance, the defects might be introduced on a material with a selection of bad quality raw material. For the yarns, the defects might come in the knitting, finishing, or dyeing processes. The defects are quite harsh as they detract from the aesthetics as well as the performance of the material. They also increase wastage that ultimately leads to customer dissatisfaction.
Computer vision systems usually involve industrialized cameras. The systems have the potential to check the work quality and give feedback for guiding production decisions. This is possible through the help of the software for image processing. Computer vision is especially important for inspecting the moving materials at great speeds up to 120 meters per minute. This is based on the production line. The ability of the system to work continuously with quality and consistency improve the bottom line of manufacturers.
Line scan cameras are in high use in the systems for detecting defects in textile manufacture. The cameras utilize the single pixels line for constructing a continual 2D image as a web material goes across the line. The cameras are good for their potential to perform with different webs of material. They help in the detection of pattern changes on a continuous note. What’s more, they also help in noting the texture and color change. They also help to note defects on different textiles that move at different predictable rates through the lines of production.
The benefit that comes with using these advanced cameras is their potential to give smear-free images at good speeds. These come with increased efficiency for processing and with lower cost for the pixels in comparison to the conventional area cameras. You can see some instances where you get high-resolution series that make the cameras potent to capture quality pictures. This helps in assisting the defect detection process.
Textile manufacturers use line scan cameras to capture images through a single light line. Advanced ones have the capacity to integrate with different LED sources of light. This is important for the detection of different LED sources for detecting defects. These may be present across the length and the width of the textile moving in the production line. It is important to have some illumination on the view field in uniform and intense ways.
We see that most of the innovative cameras might correct the non-uniformity in the illumination. However, the high intensity is important as the shorter integration times that come with line scan imaging. The shorter integration times enable the objects to go rapidly in front of the cameras. This happens without any interference of motion blurring.
The data that gets generated by the camera might be used for creating 2D images. On the other hand, it might create a map automatically that shows the location of the defects on the surface of the textile. The quality control inspector would then review the issue in the map for validating it. Some of the usual defects the inspection side looks for involve misprints, water damage, oil spots, foreign fiber, and many others.
The software for image processing analyzes the defect map or the images for constructing virtual cutting plans for inspected textile. The procedure enables the manufacturing unit to construct the cutting plan virtually. It would produce a large yield with almost no defects before the physical textile is cut. When the ideal cutting plan comes into being, the manufacturer is able to implement the plan. The textile gets prepared for the shipment as well.
It is estimated that about a total of 10 to 20 percent of the textile wastage is through production procedures. The wasted material might be counted as post-industrial waste. It might also be referred to as pre-consumer waste. The latter involves anything remaining from the production. It includes the leftovers from cutting, roll ends, and dismissed fiber material.
The 4-point system is the most important grading system to minimize the waste produced in the textile industry. Through the grading system, the fabric is given certain points. The points are based on the total width and length of the defect.
The points to be calculated include a 100 square yard fabric that is graded and rolled. Usually, this size of fabric comes with less than 40 points if it is of good quality. The rating goes the opposite otherwise.
The textile material that might fall under the threshold level is then delivered to landfills. In some cases, they are even turned to ash. However, some enterprises repurpose the textile waste into different materials for industrial use. This includes furniture, car seat, paper, and others. It might also include mattress stuffing or others. It is ideal for the manufacturers to get their better management hands on the wastage as well. This is important as it is an important way to build profit and reduce the raw material purchase. Moreover, it also offers huge protection to our environment and ecosystem.
In increasing terms, the designers for clothing now start regarding this waste as important resources to be used to make new clothes. Identifying and differentiating the clothes in terms of shape, material, color, and other things might be problematic. However, with computer vision with the help of AI and ML, the possibilities are always there.
We see how computer vision is actually an important tool that enables manufacturers in the textile industry. It helps in allowing the manufacturers to create different textiles of good quality. This would result in minimized cost and maximize return on the sale. The systems for computer visions allow the manufacturers to give virtually flawless goods. These would also help in minimising waste and promoting an environment that is sustainable. This is all possible when the defects can be measured even before the textile reaches the buyers.
In this article, we saw how computer vision-assisted visual quality inspection can be used in the textile industry. We need to develop more ways to integrate advanced quality inspection technology in the textile industry. This would not only expand the horizons of the existing market but also decrease further the load on units and the environment. We can also benefit the overall economy of the concerned country, but this is another debate.
>> Book a meeting by clicking here <<
References
https://www.sciencedirect.com/science/article/pii/B9780081012178000014
