AUTOMATED SYSTEM FOR FABRIC DESIGN SELECTION IN PATTERNED WEAVING
Keywords:
patterned textiles, automated design, deep learning, GAN, national patterns, atlas, adras, smart textileAbstract
This article proposes a system for automatic selection of fabric design (rapport, color combination, pattern placement) in the production of patterned textiles. The system consists of convolutional neural networks (CNN), generative adversarial networks (GAN) and multi-criteria decision-making algorithms. The model, trained on a special dataset of 42,000 images created based on the Uzbek national atlas and adras patterns, offers a design that meets the aesthetic requirements of the customer with 94.7% accuracy. In production tests, the design selection time was reduced from 45 minutes to 38 seconds, and the number of returned orders for color-pattern compatibility decreased by 82%.
References
Sobirova N.R. va b. (2024). O‘zbek milliy naqshlari tasnifi va raqamlashtirish. To‘qimachilik jurnali, 3(48), 12–25.
Karras T. et al. (2021). Alias-Free Generative Adversarial Networks (StyleGAN3). NeurIPS.
Zhang R. et al. (2023). ControlNet: Adding Conditional Control to Text-to-Image Diffusion Models. ICCV.
O‘zR Standarti 3278-2022. Milliy atlas va adras matolari – texnik talablar.
Xo‘jayev D.A. (2025). To‘qimachilikda sun’iy intellekt qo‘llanilishi. Monografiya, TATU nashriyoti.
Зулунов Р. Абдурасулова Д. Автоматизированная система подбора дизайнов тканей с учетом возможностей оборудования при производстве узорчатых тканей. Al-Farg'oniy avlodlari, 2025, 1(1), 83-86.
R.Zulunov, D.Abdurasulova. Ignali to‘quv dastgohining ishlash algoritmi: tuzilishi va jarayoni. Scientific-technical journal (STJ NamITI, NamTSI ITJ, НТЖ НамИТП, 2025, № 1), 19-2.
R.Zulunov, B.Soliyev, A.Kayumov, M.Asraev, Kh.Musayev, D.Abdurasulova. Detecting mobile objects with ai using edge detection and background subtraction techniques. E3S Web of Conferences, 508, 03004 (2024).
Downloads
Published
How to Cite
License
Copyright (c) 2025 Dilnoza Abdurasulova; Ravshan Zulunov
This work is licensed under a Creative Commons Attribution 4.0 International License.
