
With increasing attention to sustainable material development, bio-based materials such as bacterial cellulose (BC) have been widely explored in design research. However, their broader application remains constrained by long production cycles and limited controllability. To address these limitations, this study presents an exploratory investigation of AI-based virtual biomaterial generation under limited-data conditions. Specifically, an image dataset was first constructed from self-cultivated BC samples produced under controlled settings. Three generative adversarial network (GAN)-based models were trained and evaluated for virtual material generation. User perception experiments revealed that StyleGAN-generated samples received the highest ratings in naturalness, sense of growth, and credibility, approaching evaluations of physical BC samples. In addition, an image-to-3D proof-of-concept showed that GAN-generated textures could be used as visual inputs for preliminary 3D shape generation, while cross-material experiments demonstrated the potential for simulating novel material combinations. The findings indicate that, despite limited training data, GAN-generated textiles can capture key visual characteristics of BC and facilitate early-stage design exploration. By integrating biological experimentation with computational generation, this work proposes a AI-assisted framework with the potential to reduce physical material consumption and support sustainable design innovation. The study further suggests that generative models can function not only as visualization tools but also as assistive systems for future material design.
Fabric anomaly detection often relies on defect-free training because annotated defects are difficult to collect. Using one detector for multiple fabric categories would reduce training and maintenance costs. However, normal appearance varies considerably across weave structures, printed patterns, dyeing conditions, and local texture continuity, while many defects affect only a small region. Therefore, a unified model must adapt to the normal appearance of each fabric category while remaining sensitive to subtle local disruptions. To address these challenges, we present CALG-AD, a context-adaptive local–global reconstruction framework based on Dinomaly. Unlike Dinomaly’s directly grouped encoder targets, adaptive context decoupling constructs each reconstruction target relative to the fused class-token context and current-image spatial feature mean. Adaptive local–global attention retains Dinomaly’s global linear-attention branch and augments its value path with a routed local residual for small defects and local texture irregularities. Experiments on MVTec-AD textures, WFDD, and ZJU-Leaper evaluate the method across different fabric structures and defect shapes. On WFDD, CALG-AD improves I-AUROC from 93.41% to 97.36% and P-AUPRO from 92.58% to 93.66%. The defect-type analysis shows particularly clear P-AP gains for flecked, line, and string defects. CALG-AD also improves all three image-level metrics and P-AUPRO over Dinomaly on MVTec-AD textures and ZJU-Leaper, while requiring fewer parameters and FLOPs. These results demonstrate its effectiveness for unified anomaly detection across multiple fabric categories within each dataset. Our code is available at https://github.com/CaptaZ/CALG-AD .
Currently, the mass transfer optimization of tree-like patterns is mainly focused on mass transfer within circular channels in three-dimensional structures. In this work, a tree-like pattern for optimizing liquid water transport inside double-jersey knitted fabric is proposed for, it is believed, the first time, from both macroscopic and microscopic perspectives. Darcy’s law and the Hagen–Poiseuille equation are respectively used to investigate the optimal volume flow rate of two-dimensional tree-like patterns (with the width much larger than the thickness) under constant external pressure; this is further verified by a finite-element model based on experimental data. The results show that the liquid flow rate in the branches reaches its maximum at a fractal dimension of 1, deviating from the prediction of Murray’s law. In addition, a finite-element model with an error range of less than 5% is established based on experimental parameters and results. By comparing the liquid water transport efficiency inside the tree-like pattern under different fractal dimensions, it is found that the calculated results of the finite-element model are in good agreement with the theoretical results, providing a theoretical basis for optimizing the moisture wicking performance of smart functional fabrics and garments.
Firefighters working in high-temperature and physically demanding environments often experience sweat accumulation, which compromises wearing comfort and promotes microbial growth, creating a need for protective textiles with both reliable thermal protection and antibacterial functionality. However, conventional protective fabrics often fail to achieve antibacterial functionality while maintaining excellent flame-retardant and thermal-insulating performance, thereby limiting their overall protective efficiency. Herein, a multiscale nanofiber-embedded composite yarn was developed by using aramid 1313 and polyimide fibers as an intrinsically heat-resistant framework, followed by the electrospinning-assisted incorporation of triclosan-loaded polyacrylonitrile nanofibers in the staple-fiber yarn structure. The embedded nanofibers were designed to offer antibacterial functionality while maintaining the spinnability and weavability of the aramid and polyimide yarns. The composite yarn self-extinguished within 3 s after removal of the flame. Moreover, woven fabrics fabricated from the composite yarn generated a surface temperature difference of over 40°C compared with cotton fabrics when placed on a high-temperature hot plate, demonstrating superior thermal protective capability. In addition, the composite fabric showed favorable moisture-wicking performance and achieved antibacterial rates exceeding 99% against both Staphylococcus aureus and Escherichia coli . This work provides a feasible cross-scale nanofiber-embedding strategy for constructing multifunctional protective yarns and fabrics.
Owing to the spread of COVID-19, masks made from a variety of materials and designs have been used in the absence of guidelines. To inform measures against seasonal influenza and prepare for the next pandemic, this study conducted particle penetration and leakage tests on prototype masks to propose design guidelines for masks used in nonhealthcare settings. In Test 1, the pressure drop and permeability were measured for six types of flat fabric. In Test 2, permeability measurements were performed on 12 prototype masks made from three different materials and categorized into four sizes. The masks were fitted onto facial dummy designed to match the contours of the human face. The peripheral leakage rate was calculated based on the measured leakage. Furthermore, an empirical equation was formulated to estimate the pressure drop Δ P w during wear using the ratio of the flow velocities in each test and the pressure drop of the flat fabric. Subsequently, the peripheral and total leakage rates were plotted against Δ P w ; the transition of leakage from the gaps to the part that covered the nose and mouth occurred at Δ P w = 1 Pa. Next, using the percentage of the peripheral leakage rate to the total leakage rate at 1Pa as an exploratory indicator in the empirical model, the minimum material area required to minimize peripheral leakage was estimated. This estimation method determines the required material area requirements based on the material properties. These findings support the development of standardized methods for mask design and analysis.