A Method for the Assessment of Textile Pilling Tendency Using Optical Coherence Tomography

Prediction of textile pilling resistance using optical coherence tomography

This paper describes a new method of textile pilling prediction, based on multivariate analysis of the spatial layer above the surface. The original idea of the method is the acquisition of 3D fabric image using optical coherence tomography (OCT) with infrared light, which allows for the fabric fuzz visualization with high sensitivity. The pilling layer, reconstructed with the resolution of [Formula: see text], includes reliable textural information related to the amount of loose fibers and bunches appearing as a result of abrasion. Pilling intensity was assigned by supervised classification of the textural features using both linear (PLS-DA - partial least squares discriminant analysis, LDA - linear discriminant analysis) and non-linear (SVM - support vector machine) classifiers. The results demonstrated that the method is more suitable for fabrics after short-term abrasion, when the fuzz prevails over tangled fibers in the pilling layer. In that case, pilling grades were predicted with [Formula: see text] accuracy, sensitivity and specificity (for SVM model). The validation accuracy of the tested models after machine abrasion achieves lower values (up to [Formula: see text] for LDA model). With our method, we clearly showed that OCT can be used to quantitatively trace appearance changes of fabric samples due to test abrasion.

Comparison of Machine Learning Methods for Image Reconstruction Using the LSTM Classifier in Industrial Electrical Tomography

Electrical tomography is a non-invasive method of monitoring the interior of objects, which is used in various industries. In particular, it is possible to monitor industrial processes inside reactors and tanks using tomography. Tomography enables real-time observation of crystals or gas bubbles growing in a liquid. However, obtaining high-resolution tomographic images is problematic because it involves solving the so-called ill-posed inverse problem. Noisy input data cause problems, too. Therefore, the use of appropriate hardware solutions to eliminate this phenomenon is necessary. An important cause of obtaining accurate tomographic images may also be the incorrect selection of algorithmic methods used to convert the measurements into the output images. In a dynamically changing environment of a tank reactor, selecting the optimal algorithmic method used to create a tomographic image becomes an optimization problem. This article presents the machine learning method’s original concept of intelligent selection depending on the reconstructed case. The long short-term memory network was used to classify the methods to choose one of the five homogenous methods—elastic net, linear regression with the least-squares learner, linear regression with support vector machine learner, support vector machine model, or artificial neural networks. In the presented research, tomographic images of selected measurement cases, reconstructed using five methods, were compared. Then, the selection methods’ accuracy was verified thanks to the long short-term memory network used as a classifier. The results proved that the new concept of long short-term memory classification ensures better tomographic reconstructions efficiency than imaging all measurement cases with single homogeneous methods.

Changes in Fabric Surface Pilling under Laser Ablation

Textiles require finishing to improve their usability and functionality but in the first place, to reduce of pilling tendency, which affects all kinds of synthetic and natural fabrics. Several laser ablation tests have been applied to the selected fabrics with different chemical composition to reveal the impact of this process on the pilling behavior. To reflect the pilling changes, two textural descriptors have been proposed to textile images obtained with optical coherence tomography (OCT). They showed the trend to reduce values with increasing laser power applied to the tested fabrics. It has been demonstrated, that in the case of textiles based on polyester threads, laser modification of the product surface led to a significant reduction in their tendency for pilling.