علوم و فناوری نساجی و پوشاک

علوم و فناوری نساجی و پوشاک

ارائه چارچوبی داده‌محور برای طبقه‌بندی و مشابهت‌سنجی فرش‌های ایرانی با الگوریتم K-Means

نوع مقاله : مقاله پژوهشی

نویسنده
مهران دادگر
گروه مهندسی نساجی، دانشکده فنی و مهندسی ، دانشگاه نیشابور، نیشابور، ایران
10.22034/jtst.2025.531101.1504
چکیده
این پژوهش با هدف طراحی یک سامانه عینی طبقه‌بندی برای فرش‌های اصیل ایرانی انجام شد. یافته‌های تحقیق مبنایی برای توسعه سامانه جهانی طبقه‌بندی فرش فراهم می‌کند که می‌تواند به ایجاد شاخص کمّی ارزش‌گذاری و سامانه گواهی‌نامه‌دهی یکپارچه منجر شود. در این راستا، پایگاه‌داده‌ای جامع از ۹۹ نمونه فرش موزه‌ای گردآوری شد که شامل 12 پارامتر مرتبط با ساختار، طرح و مواد اولیه بود. برای دستیابی به این اهداف، الگوریتم خوشه‌بندی ک. مین با فاصله اقلیدسی و 3k= بر اساس روش کمانی به‌کار گرفته شد و نمونه‌ها بر اساس تراکم گره، ترکیب مواد و جنس پرز به سه خوشه همگن تقسیم شدند. نوآوری اصلی این تحقیق در آن است که با تبدیل معیارهای سنتی و ذهنی به شاخص‌های عینی، امکان مقایسه مستقیم فرش‌های نو با نمونه‌های تاریخی فراهم می‌شود. این نظام نه تنها دقت در ارزش‌گذاری را افزایش می‌دهد، بلکه می‌تواند به شفافیت قیمت‌گذاری، توسعه بازارهای جهانی، اعتماد بیشتر خریداران، فراهم نمودن امکان خرید خریدار بر طبق ذائقه و علاقه خود به فرش یک منطقه خاص و فراهم کردن بستر اعتماد به نظر فروشندگان منجر گردد.
کلیدواژه‌ها
خوشه
تایید دسته بندی
ارزیابی
گره
فرش
موضوعات

عنوان مقاله English

A Data-Driven Framework for Classification and Similarity Assessment of Persian Carpets Using K-Means Algorithm

نویسنده English

Mehran Dadgar
Department of Textile Engineering, Faculty of Engineering, Neishabour University, Neishabour, Iran
چکیده English

This study was conducted with the aim of designing an objective classification system for authentic Iranian carpets. The research findings provide a basis for developing a global carpet classification system, which could lead to the establishment of a quantitative valuation index and an integrated certification system. In this regard, a comprehensive database of 99 museum carpet samples was collected, containing 12 parameters related to structure, design, and raw materials. To achieve these objectives, K-means clustering algorithm with Euclidean distance and k=3 was employed based on the elbow method, and the samples were divided into three homogeneous clusters based on knot density, material composition, and pile material. The main innovation of this research lies in converting traditional and subjective criteria into objective indices, enabling direct comparison of new carpets with historical samples. This system not only increases valuation accuracy but can also lead to pricing transparency, development of global markets, greater buyer confidence, the possibility for buyers to purchase according to their taste and interest in carpets from a specific region, and providing a basis for trust in sellers' opinions.

کلیدواژه‌ها English

clustring
verification
valuation
knot
carpet
1.    Bassam, J., Bilingual culture of handwoven carpet. 2012, Tehran: Encyclopedia Foundation.
2.   Tajadod, H., Encyclopaedia of Iranian handwoven carpets. 2011, Tehran: Encyclopaedia of Iranology.
3.   Yavari, H., The basics of Iranian carpet recognition. 2005, Tehran: Raja Publications.
4.   Tabatabaei, S.M., et al., Modeling the Performance Properties on Woolen Handknotted Carpets using Response Surface Methodology. Journal of Fashion Technology & Textile Engineering, 2016. 2014.
5.   Serrano, A., et al., A non-invasive imaging approach for improved assessments on the construction and the condition of historical knotted-pile carpets. Journal of Cultural Heritage, 2021. 47: p. 79-88.
6.   Kashyzadeh, K.R. and A.H. Darounkola, Application of Taguchi approach to forecast the wages of persian silk carpet weavers. International Journal of Engineering and Management Research, 2021. 11.
7.   Baranwal, B., Classification of carpets, in Advances in Carpet Manufacture (Second Edition), K.K. Goswami, Editor. 2018, Woodhead Publishing. p. 467-483.
8.   Elkhateeb, A., et al., Absorption characteristics of masjid carpets. Applied Acoustics, 2016. 105: p. 143-155.
9.   ÖZDİL, N., et al., COMPRESSIBILITY AND THICKNESS RECOVERY CHARACTERISTICS OF CARPETS. Journal of Textile & Apparel/Tekstil ve Konfeksiyon, 2012. 22(3).
10. Dadgar, M., et al., Stability of Polypropylene Frieze Yarn after Different Heat Setting Condition and Prediction Residual Frieze Effect By Mamdani's Fuzzy Inference System, in AUTEX. 2013: Germany.
11. Dadgar, M., Prediction of Polypropylene Yarn Shrinkage in Heat Setting Process Using Fuzzy Inference System. FIBRES & TEXTILES in Eastern Europe 2020.
12. Dadgar, M., S.M. Hosseini Varkiyani, and A.A. Merati, Pin-Point Effect Determination Using A Rigorous Approach. Indian Journal of Textile, 2014.
13. Dadgar, M., S.M. Hosseini Varkiyani, and A.A. Merati, A New method of Evaluating Heat Setting Effect of Carpet Yarns, in ATC. 2011: Korea.
14. Dadgar, M., A.A. Merati, and S.M. Hosseini Varkiyani, Evaluation of the pinpoint effect on carpet appearance. The Journal of The Textile Institute, 2015: p. 1-9.
15. Dadgar, M., S.M. Hosseini Varkiyani, and A.A. Merati, Evaluation of heat setting parameters in Carpet Comfort. Fiber and polymer, 2015.
16. Dadgar, M., Developing new measuring and ANN prediction method for the longitudinal diameter of intermingled untwisted heat set bulk continues filament yarns. Measurement, 2021. 171: p. 108727.
17. Dadgar, M., et al., Comparison of Mamdani and Sugeno fuzzy inference system in prediction of residual frieze effect of frieze carpet yarns. Tekstilna industrija, 2013. 61(2): p. 16-25.
18. Dadgar, M. and S.M. Hosseini Varkiyani, Comparison heat setting results of SUPERBA and POWER HEAT SET. Tekstilna industrija, 2014.
19. Dadgar, M., S.M. Hosseini Varkiyani, and A.A. Merati, Comparison between artificial neural network and response surface methodology in the prediction of the parameters of heat set polypropylene yarns. The Journal of The Textile Institute, 2014. 106(4): p. 417-430.
20. Dadgar, M. and K. Noshad, Machine Learning Modeling of Melt-Spinning for Yarn Property Prediction. Results in Materials, 2025: p. 100776.
21. Abidi, F., et al., A General System of Handmade Carpet Geometry. International Journal of Modern Engineering Research, 2015. 5(11): p. 72-76.
22. Fatma, A., et al., PREDICTION OF THE HANDMADE CARPET WEIGHT. 2014.
23. Abidi, F., et al., Modelling of Tunisian hand-made carpet resilience under long static loading using fuzzy expert system. Fibers and Polymers, 2017. 18(9): p. 1810-1815.
24. Gupta, S.K., K.K. Goswami, and A. Majumdar, Durability of Handmade Wool Carpets: A Review. Journal of Natural Fibers, 2015. 12(5): p. 399-418.
25. Kakuee, O., et al., Investigation of mordants for dyeing of yarns in ancient Persian carpets (15th–17th century) by IBA methods. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, 2019. 450: p. 294-298.
26. Barbache, S., et al., Optical analyses of wool dyeing materials in ancient Moroccan carpets “Zarbia (s)”: Combination of UV-vis diffuse reflectance, 3D-fluorescence and Raman spectroscopies. Dyes and Pigments, 2018. 153: p. 256-265.
27. Hasanov, E.L., Innovative basis of research of technologic features of some craftsmanship traditions of Ganja (On the sample of carpets of XIX century). International Journal of Environmental and Science Education, 2016. 11(14): p. 6704-6714.
28. Manhita, A., et al., Unveiling the colour palette of Arraiolos carpets: Material study of carpets from the 17th to 19th century period by HPLC-DAD-MS and ICP-MS. Journal of cultural heritage, 2014. 15(3): p. 292-299.
29. Csapó, J., et al., Age estimation of old carpets based on cystine and cysteic acid content. Analytica chimica acta, 1995. 300(1-3): p. 313-320.
30. Lodewijks, J., The use of synthetic material for the conservation and restoration of ancient textiles. Studies in Conservation, 1964. 9(sup1): p. 79-85.
31. Ahmed, H., Conservation Strategies of a Rare Historical Carpet in Cairo University, A case study. Int J Fashion Technol Textile Eng, 2018. 1(1): p. 1-6.
32. Tousian Shandiz, G., et al., Pathology and Classification of Iranian CarpetDamages. Journal Scientific Goljaam, 2014. 9(24): p. 55-70.
33. Petillo, E. Comparison of treatments for damage in Oriental carpets. in Textile Specialty Group postprints: Vol. Two. Papers delivered at the Textile Subgroup Session, AIC 20th annual meeting, Buffalo, New York, June 1992. 1992.
34. Wilkinson, T.J., The definition of ancient manured zones by means of extensive sherd-sampling techniques. Journal of Field Archaeology, 1982. 9(3): p. 323-333.
35. Duan, Q. and L. Zheng, About qushu: Carpets or Rugs with Long Hair, in Studies on the History and Culture Along the Continental Silk Road. 2020, Springer. p. 21-30.
36. Huseynov, M., ANCIENT BELIEFS AS THE SOURCE OF DESIGN CHARACTERISTICS IN AZERBAIJANI CARPETS. Paradigmata poznani, 2020(4): p. 43-45.
37. Birdwood, G., THE ANTIQUITY OF ORIENTAL CARPETS. Journal of the Royal Society of Arts, 1908. 56(2920): p. 1041-1059.
38. Stead, R., The Ardabil Carpets. 1974: Getty Publications.
39. Eiland, M.L., The past re-made: the case of oriental carpets. Antiquity, 1993. 67(257): p. 859-863.
40. Singer, C., Hand-woven Carpets: Oriental and European. Nature, 1923. 111(2777): p. 71-72.
41. Manitta, S.E., The care of rugs and carpets: the case for conservation, in Restoration: is it acceptable? p. 95-101.
42. Quataert, D., MACHINE BREAKING AND THE CHANGING CARPET INDUSTRY OF WESTERN ANATOLIA 1860-1908, in Workers, Peasants and Economic Change in the Ottoman Empire, 1730-1914. 2010, Gorgias Press. p. 117-136.
43. Zeinalpour Yazdi, M. and P. Pevandi, A review of camouflage pattern generation, evaluation, and detection methods. Textile and Clothing Science and Technology, 2017. 6(3): p. 9-21.
44. Zare Nejad, Z. and P. Pevandi, Clothing Image Clustering Using Image Processing and K-Means Algorithm. Textile and Clothing Science and Technology, 2013. 3(1): p. 3-10.
45. Company, I.C.J.S., Recognizing examples of Iranian carpets. Vol. 1. 1982, Tehran,IRAN: Iran Carpet Company. 220.
46. Company, F.I.J.S., Knowing examples of Iranian carpets. 1361, Tehran: Pishgar.
47. Soltani, P., et al., A study on siro, solo, compact, and conventional ring-spun yarns. Part III: modeling fiber migration using modular adaptive neuro-fuzzy inference system. Journal of The Textile Institute, 2013(ahead-of-print): p. 1-11.
48. Anderberg, M., Cluster analysis for applications. 1973, New York: Academic Press, Inc.
49. Zhang, S., et al., Phenomapping of subgroups in high-Lp (a) patients: a data-driven cluster analysis in RED-CARPET study. Clinical Research in Cardiology, 2025: p. 1-11.
50. Sambade, I.M., et al., Signatures of Selection for Resistance/Tolerance to Perkinsus olseni in Grooved Carpet Shell Clam (Ruditapes decussatus) Using a Population Genomics Approach. Evolutionary Applications, 2025. 18(5): p. e70106.
51. Cutri, R., O. Mattasoglio, and N. Stem. Data Mining Application in an Introductory Engineering Physics Lab. in 2025 ASEE Annual Conference & Exposition. 2025.
52. Cutri, R., N. Stem, and O.M. Neto, Data mining application in an introductory engineering physics lab: a diagnostic activity. Brazilian Journal of Development, 2025. 11(3): p. e78196-e78196.
53. Mirjalili, R., et al. VLM-Vac: Enhancing Smart Vacuums Through VLM Knowledge Distillation and Language-Guided Experience Replay. in 2025 IEEE International Conference on Robotics and Automation (ICRA). 2025. IEEE.
54. Sabillon Orellana, C.A., et al., Pavement Patterns: An Unsupervised Learning Approach to Texture Analysis in Rigid Pavements. Transportation Research Record, 2025: p. 03611981251355527.
55. Amirian Varnousefaderani, B., et al., Enforced clustering for zero-to-one-shot texture anomaly detection. Machine Vision and Applications, 2025. 36(3): p. 53.

  • تاریخ دریافت 12 تیر 1404
  • تاریخ بازنگری 26 مهر 1404
  • تاریخ پذیرش 26 مهر 1404