Textile Technology

Textile technology spans the design, production and functionalisation of fibres, yarns and fabrics, integrating advances in materials science, engineering and data-driven methods. From natural cellulosic and protein fibres to synthetic polymers, innovations in fibre chemistry and processing underpin performance improvements in apparel, industrial and technical textiles. Spinning technologies now combine precise tension control, twist‐generation modelling and compacting elements to deliver yarns with enhanced strength, uniformity and specialised functions. Fabric formation leverages weaving, knitting and non‐woven techniques, often supplemented by in‐line sensors and robotic handling for efficient patterning and quality assurance. Simultaneously, sustainable finishing methods—ranging from supercritical CO₂ dyeing to enzyme-mediated natural mordanting—are reducing water, energy and chemical footprints. Across the value chain, automated inspection using deep learning, coupled with nanocomposite coatings and biodegradable crosslinkers, is enabling next-generation textiles that marry wear comfort, mechanical resilience and environmental responsibility.

Research from Nature Portfolio

Recent studies have advanced automated quality control by integrating density-based anchor clustering with a dual-channel feature enhancement module atop a dense convolutional backbone. This system localises minor surface flaws in real time at nearly 99% mean average precision while maintaining 40 frames per second throughput. In parallel, supercritical carbon dioxide has been harnessed for water-free textile finishing: novel azo-pyrazole disperse dyes deliver high colour strength and fastness on polyester, whereas organoclay-assisted scCO₂ dyeing of polypropylene nanocomposites enhances dye uptake and wash durability. Together, these developments demonstrate the convergence of green chemistry and deep learning for inline defect detection and solvent-free functionalisation of synthetic fabrics.

Research from all publishers

Non-Portfolio research continues to refine machine-vision inspection tools. A mask-guided diffusion augmentation framework learns fabric texture–defect correlations to generate synthetic flaw samples, and attention-guided fusion networks double detection sensitivity while halving model size. Unsupervised multi-scale denoising autoencoders reconstruct defect-free image patches across Gaussian pyramids, flagging anomalies via reconstruction residuals and eliminating the need for defective training data. Another line of work enhances single-stage detectors with spatial pooling and adaptive histogram equalisation, boosting mean average precision by over 10% under industrial throughput constraints. These lightweight architectures exemplify the drive toward rapid, accurate fabric inspection with minimal computational overhead.

Textile Technology publication trend

The graph below shows the total number of articles in textile technology across all publications each year (not limited to Nature Index journals).

Technical terms

Supercritical carbon dioxide (scCO₂): A state of CO₂ above its critical temperature and pressure, combining liquid-like solvency with gas-like diffusivity for water-free dyeing or extraction.

Diffusion model: A generative approach that learns data distributions by iteratively adding and removing noise, here used to synthesise realistic defect examples.

Denoising autoencoder: An unsupervised neural network trained to reconstruct clean inputs from noisy versions, identifying anomalies via reconstruction error.

Attention mechanism: A network module that adaptively weights feature channels or spatial regions, enabling focus on salient defect characteristics.

Mean average precision (mAP): A performance metric averaging precision values across recall levels for object detection, summarising localisation and classification accuracy.

References

1. Fabric defect detection using the improved YOLOv3 model. Journal of Engineered Fibers and Fabrics (2020).
2. Automatic Fabric Defect Detection with a Multi-Scale Convolutional Denoising Autoencoder Network Model. Sensors (2018).
3. Research on fabric surface defect detection algorithm based on improved Yolo_v4. Scientific Reports (2024).
4. Synthesis of novel azo pyrazole disperse dyes for dyeing and antibacterial finishing of PET fabric under supercritical carbon dioxide. Scientific Reports (2024).
5. Organoclay-assisted disperse dyeing of polypropylene nanocomposite fabrics in supercritical carbon dioxide. Scientific Reports (2024).
6. Textile and Fabric Manufacture.