信息与控制2025，Vol.54Issue(4)：658-672,15.DOI:10.13976/j.cnki.xk.2024.1082
基于周期性细粒度纹理生成网络的碳纤维复合材料纹理图像生成
Texture Image Generation for Carbon Fiber-Reinforced Composites Based on Periodic Fine-grained Texture Generation Network
摘要
Abstract
To address the issue of insufficient texture image data during the quality inspection of Carbon Fiber-Reinforced Composites(CFRCs)layup process,we propose a periodic fine-grained texture generation network(PFTN)and a defect transfer PFTN(DT-PFTN)with defect transfer function-ality.The PFTN effectively balances the randomness and periodicity of texture images by introducing a fine-grained texture compensation module and a learnable periodic sine wave,ensuring the gen-eration of high-quality,fine-grained CFRCs texture images.The defect transfer network DT-PFTN generates fiber texture images with real surface defects by effectively encoding and embedding de-fect features.We contruct a CFRCs texture image dataset using the proposed models,including 12 480 normal texture image samples and 3 000 defective texture image samples.Experimental re-sults demonstrate that both the PFTN and DT-PFTN models achieve competitive texture image gen-eration effects on both defect-free and defective datasets,with SIFID(Single Image Fréchet Incep-tion Distance)(10-6),LPIPS(Learned Perceptual Image Patch Similarity),and FID(Fréchet Inception Distance)scores reaching 9,0.35 and 60.81,respectively.关键词
纹理图像生成/碳纤维复合材料/数据增强Key words
texture generation/carbon fiber composites/data augmentation分类
信息技术与安全科学引用本文复制引用
许原野,张吟龙,梁炜..基于周期性细粒度纹理生成网络的碳纤维复合材料纹理图像生成[J].信息与控制,2025,54(4):658-672,15.基金项目
国家自然科学基金项目(62273332) （62273332）
中国科学院青年创新促进会会员项目(202201) （202201）
广东省基础与应用基础研究基金(2023A151501136) （2023A151501136）
