The process of dyeing fabric can lead to textile waste due to the typical method of coloring while wet, which makes it difficult to know how the color will look once dry. But a professor at North Carolina State University’s Wilson College of Textiles has found a way to solve this problem using machine learning.
Professor Warren Jasper found that the amount of color change from wet to dry states is not uniform between different colors, and that non-linear relationship makes it difficult to anticipate results based on one color sample.
Related Stories
“The fabric is dyed while wet, but the target shade is when its dry and wearable. That means that, if you have an error in coloration, you aren’t going to know until the fabric is dry,” he said. “While you wait for that drying to happen, more fabric is being dyed the entire time. That leads to a lot of waste, because you just can’t catch the error until late in the process.”
To solve this problem, Jasper developed five machine learning models, including a neural network designed to map this kind of non-linear relationship. Using visual data from 763 fabric samples of various colors in both wet and dry states, Jasper trained the machines to anticipate results.
Each of the machine-learning models outperformed those not using AI in terms of accuracy, but the neural network surpassed all the others with an error value as low as .01 and a median error of 0.7 using CIEDE2000, a standardized color difference formula. The other machine learning models showed CIEDE2000 error ranges betwen 1.1 and 1.6, with a baseline as high as 13.8. Error values exceeding 0.8 to 1 are generally considered outside acceptable limits in the textile industry.
Jasper found that the neural network has the potential to significantly cut waste caused by color errors, since it would allow fabric makers to better predict the end result of dyeing before large amounts of textiles have been incorrectly colored. He outlined his findings in the paper, “A Controlled Study on Machine Learning Applications to Predict Dry Fabric Color from Wet Samples: Influences of Dye Concentration and Squeeze Pressure,” published in the journal Fibers.
Machine learning and AI are being tapped by the textile industry in other sectors, primarily recycling and circularity. Jasper said he hopes this research will lead to wider use of similar machine-learning tools across the broader textile industry.
“We’re a bit behind the curve in textiles. The industry has started to move more toward machine learning models, but it’s been very slow,” he said. “These types of models can offer powerful tools in cutting down on waste and improving productivity in continuous dyeing, which accounts for over 60 percent of dyed fabrics.”
