Saved in:
Bibliographic Details
Main Authors: Kainz, Maria, Krondorfer, Johannes K., Jaschik, Malte, Jernej, Maria, Ganster, Harald
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.03575
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910929691607040
author Kainz, Maria
Krondorfer, Johannes K.
Jaschik, Malte
Jernej, Maria
Ganster, Harald
author_facet Kainz, Maria
Krondorfer, Johannes K.
Jaschik, Malte
Jernej, Maria
Ganster, Harald
contents Recycling textile fibers is critical to reducing the environmental impact of the textile industry. Hyperspectral near-infrared (NIR) imaging combined with advanced deep learning algorithms offers a promising solution for efficient fiber classification and sorting. In this study, we investigate supervised and unsupervised deep learning models and test their generalization capabilities on different textile structures. We show that optimized convolutional neural networks (CNNs) and autoencoder networks achieve robust generalization under varying conditions. These results highlight the potential of hyperspectral imaging and deep learning to advance sustainable textile recycling through accurate and robust classification.
format Preprint
id arxiv_https___arxiv_org_abs_2505_03575
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Supervised and Unsupervised Textile Classification via Near-Infrared Hyperspectral Imaging and Deep Learning
Kainz, Maria
Krondorfer, Johannes K.
Jaschik, Malte
Jernej, Maria
Ganster, Harald
Computer Vision and Pattern Recognition
Applied Physics
Recycling textile fibers is critical to reducing the environmental impact of the textile industry. Hyperspectral near-infrared (NIR) imaging combined with advanced deep learning algorithms offers a promising solution for efficient fiber classification and sorting. In this study, we investigate supervised and unsupervised deep learning models and test their generalization capabilities on different textile structures. We show that optimized convolutional neural networks (CNNs) and autoencoder networks achieve robust generalization under varying conditions. These results highlight the potential of hyperspectral imaging and deep learning to advance sustainable textile recycling through accurate and robust classification.
title Supervised and Unsupervised Textile Classification via Near-Infrared Hyperspectral Imaging and Deep Learning
topic Computer Vision and Pattern Recognition
Applied Physics
url https://arxiv.org/abs/2505.03575