Design of Miao Embroidery Gene Decoding and Digital Activation Based on CNN-Transformer
Downloads
To overcome the challenges of decoding pattern genes and the lack of innovative designs in the digital transformation of Miao embroidery as an intangible cultural heritage, a deep learning fusion architecture based on a convolutional neural network Transformer (CNN Transformer) is proposed. This architecture aims to accurately analyze the semantics and generate stylized Miao embroidery patterns. This architecture uses a ResNet-50 backbone network to extract local pattern details and combines an improved atrous spatial pyramid pooling (ASPP) module to preserve edge structures for semantic segmentation. It simultaneously utilizes a progressive Transformer encoder and multi-scale channel attention to dynamically focus on the pattern subject, fuse features, and achieve style transfer. The experimental results showed that the segmentation accuracy of the model on four types of embroidery patterns exceeded 85% (up to 92.68%), and the structural similarity (SSIM) reached 0.957. In the style transfer task, the style fidelity index (KL divergence 0.52, Gram matrix MSE 0.019) and average accuracy (AP 0.926) of this method were significantly better than the comparison algorithms. This research method effectively solves the problems of inaccurate segmentation of complex patterns and distortion of cultural symbols in traditional style transfer methods. It provides a technical reference for analyzing and innovatively activating Miao embroidery culture.
Downloads
[1] Chen, L., Su, Z., He, X., Chen, X., & Dong, L. (2022). The application of robotics and artificial intelligence in embroidery: challenges and benefits. Assembly Automation, 42(6), 851–868. doi:10.1108/AA-07-2022-0183.
[2] Han, D., & Cong, L. (2023). Miao traditional patterns: the origins and design transformation. Visual Studies, 38(3–4), 425–432. doi:10.1080/1472586X.2021.1940261.
[3] Chen, L., Daud, W. S. A. W. M., & Arif, M. F. M. (2024). Preservation and Adaptation of Traditional Miao (Hainan) Pattern Arts: Themes, Shapes, and Colors in Contemporary Art. Asian Journal of Research in Education and Social Sciences, 6(4), 269-284.
[4] Wenji, Z., Rongrong, C., & Li, N. (2022). Design and Cultural Aspects of 20th Century Chinese Xiangjin Brocade. Fibres & Textiles in Eastern Europe, 151(3), 116–129. doi:10.2478/ftee-2022-0030.
[5] Qiao, S., & Abdramanov, Y. (2024). Exploratory sequential design for creation of the Miao’s traditional patterns mobile application. Culture and History Digital Journal, 13(2), 286. doi:10.3989/chdj.2024.286.
[6] Zhu, S., & Liu, X. (2025). The Ecodesign Transformation of Smart Clothing: Towards a Systemic and Coupled Social–Ecological–Technological System Perspective. Sustainability (Switzerland), 17(5), 2102. doi:10.3390/su17052102.
[7] Ni, M., Huang, Q., Ni, N., Zhao, H., & Sun, B. (2024). Research on the Design of Zhuang Brocade Patterns Based on Automatic Pattern Generation. Applied Sciences (Switzerland), 14(13), 5375. doi:10.3390/app14135375.
[8] Ji, J., Lao, Y., & Huo, L. (2024). Convolutional neural network application for supply–demand matching in Zhuang ethnic clothing image classification. Scientific Reports, 14(1), 13348. doi:10.1038/s41598-024-64082-9.
[9] Zhao, Q. (2025). Algorithm Optimization and Innovative Design of Digital Inheritance System for Miao Family Weaving Handicraft in South Sichuan Province. Journal of Combinatorial Mathematics and Combinatorial Computing, 127b, 7171–7195. doi:10.61091/jcmcc127b-391.
[10] Liu, C., Gu, J., Yao, L., & Zhang, Y. (2025). Research on embroidery style migration model based on texture cycle GAN. International Journal of Clothing Science and Technology, 37(1), 138–153. doi:10.1108/IJCST-04-2023-0062.
[11] Fan, T., Wang, H., & Hodel, T. (2023). CICHMKG: a large-scale and comprehensive Chinese intangible cultural heritage multimodal knowledge graph. Heritage Science, 11(1), 115. doi:10.1186/s40494-023-00927-2.
[12] Du, D., Ding, J., & Liu, Y. (2025). Knowledge graph construction of Chinese embroidery evolution based on associating cultural space and critical incidents under intangible cultural heritage. Electronic Library, 43(3), 283–302. doi:10.1108/EL-02-2024-0036.
[13] Xu, L., Lu, L., Liu, M., Song, C., & Wu, L. (2024). Nanjing Yunjin intelligent question-answering system based on knowledge graphs and retrieval augmented generation technology. Heritage Science, 12(1), 118. doi:10.1186/s40494-024-01231-3.
[14] Ai, L. (2026). Decoding Color: A Three-Stage Clustering Approach to Analysis Color Evolution in Changning Miao Costumes. Color Research & Application, 51(1), e70029. doi:10.1002/col.70029.
[15] Zhao, D., & Xin, W. (2025). Research on redesign of Chinese local embroidery patterns based on extension semantics and the pix2pix model. Textile Research Journal , 95(21–22), 2691–2706. doi:10.1177/00405175251314903.
[16] Yu, Q., & Zhu, G. (2023). Digital Restoration and 3D Virtual Space Display of Hakka Cardigan Based on Optimization of Numerical Algorithm. Electronics (Switzerland), 12(20), 4190. doi:10.3390/electronics12204190.
[17] Xu, M., Tao, X., Liu, J., Pan, L., & Huang, Z. (2025). Innovation in Wickerwork Forms through the Integration of Plural Geometry and AI Algorithms, and Methods for Preserving the Cultural Symbolism of the Miao People’s Flower Bamboo Hats. International Journal for Housing Science and Its Applications, 46(3), 6696–6707. doi:10.70517/ijhsa463575.
[18] Kang, X., You, W., & Xie, H. (2025). An innovative and sustainable design of intangible Miao wax printing patterns in combination of diffusion model and fuzzy TOPSIS. Humanities and Social Sciences Communications, 12(1), 1–16. doi:10.1057/s41599-025-05724-9.
[19] Zhong, C., Yu, X., Xia, H., Xie, R., & Xu, Q. (2025). Restoring intricate Miao embroidery patterns: a GAN-based U-Net with spatial-channel attention. Visual Computer, 41(10), 7521–7533. doi:10.1007/s00371-025-03821-z.
[20] Ju, F., Wang, Q., Tan, Z., & Li, Q. (2022). Intelligent Recognition of Colour and Contour from Ancient Chinese Embroidery Images. Fibres & Textiles in Eastern Europe, 151(3), 79–92. doi:10.2478/ftee-2022-0026.
[21] Zhang, Y., Zhao, H., Qi, L., Zhang, J., & Zhang, T. (2025). Research on the co-occurrence feature mining of the Qing Dynasty embroidery patterns based on temporal multilayer networks. NPJ Heritage Science, 13(1), 228. doi:10.1038/s40494-025-01766-z.
[22] Na, Z., & Sharudin, S. A. (2024). Research on Innovative Development of Miao Embroidery Intangible Cultural Heritage in Guizhou, China Based on Digital Design. GATR Journal of Business and Economics Review, 9(2), 85–94. doi:10.35609/jber.2024.9.2(1).
[23] Cong, X., & Zhang, W. (2024). Design of geometric flower pattern for clothing based on deep learning and interactive genetic algorithm. Journal of Intelligent Systems, 33(1), 57–76. doi:10.1515/jisys-2023-0269.
[24] Qian, M., Wang, Z., Huang, X., Xiang, Z., Wei, P., & Hu, X. (2022). Color segmentation of multicolor porous printed fabrics by conjugating SOM and EDSC clustering algorithms. Textile Research Journal, 92(19–20), 3488–3499. doi:10.1177/00405175221083214.
[25] Chiu, M. C., Chiang, Y. H., & Chiu, J. E. (2024). Developing an explainable hybrid deep learning model in digital transformation: an empirical study. Journal of Intelligent Manufacturing, 35(4), 1793–1810. doi:10.1007/s10845-023-02127-y.
[26] Zhao, Y., Fan, Z., Yao, H., Zhang, T., & Seng, B. (2025). Automatic Classification and Recognition of Qinghai Embroidery Images Based on the SE-ResNet152V2 Model. IET Image Processing, 19(1), 70108. doi:10.1049/ipr2.70108.
[27] Cho, H.-Y. (2021). The Language of Miao Embroidery: Exploring the Traditional “Embroidered Rear Skirt Panels” Worn by the Miao Women of the Huawu Village. TEXTILE, 21(1), 2–31. doi:10.1080/14759756.2021.1959821.
[28] Alaeddine, H., & Jihene, M. (2021). Deep residual network in network. Computational Intelligence and Neuroscience, 2021(1), 6659083. doi:10.1155/2021/6659083.
[29] Hu, Q., Peng, Y., Xu, J., Shao, Z., Tian, Z., & Chen, J. (2025). Adaptive Stylized Image Generation for Traditional Miao Batik Using Style-Conditioned LCM-LoRA Enhanced Diffusion Models. Mathematics, 13(12), 1947. doi:10.3390/math13121947.
[30] Hou, X., Zhao, H., & Wang, C. (2024). Hierarchical segmentation for traditional cultural pattern based on iterative compression and clustering. Multimedia Systems, 30(6), 372. doi:10.1007/s00530-024-01578-4.
[31] Zhou, Y., Ren, Y., Wu, C., & Xue, M. (2024). Style Transfer of Chinese Wuhu Iron Paintings Using Hierarchical Visual Transformer. Sensors, 24(24), 8103. doi:10.3390/s24248103.
[32] Zhuo, X., Huang, D., Lin, Y., & Huang, Z. (2024). Combined query embroidery image retrieval based on enhanced CNN and blend transformer. Scientific Reports, 14(1), 27518. doi:10.1038/s41598-024-79012-y.
[33] He, W., Song, B., Zhang, N., Xiang, J., & Pan, R. (2024). Modeling and realization of image-based garment texture transfer. Visual Computer, 40(9), 6063–6079. doi:10.1007/s00371-023-03153-w.
[34] Sheng, H., Cai, S., Zheng, X., & Lau, M. (2025). Knitting Robots: A Deep Learning Approach for Reverse-Engineering Fabric Patterns. Electronics (Switzerland), 14(8), 1605. doi:10.3390/electronics14081605.
[35] Fang, B., Jiang, M., Shen, J., & Stenger, B. (2022). Deep Generative Inpainting with Comparative Sample Augmentation. Journal of Computational and Cognitive Engineering, 1(4), 174–180. doi:10.47852/bonviewJCCE2202319.
[36] Wojciuk, M., Swiderska-Chadaj, Z., Siwek, K., & Gertych, A. (2024). Improving classification accuracy of fine-tuned CNN models: Impact of hyperparameter optimization. Heliyon, 10(5), 26586. doi:10.1016/j.heliyon.2024.e26586.
- This work (including HTML and PDF Files) is licensed under a Creative Commons Attribution 4.0 International License.
