Font Size: a A A

The Research On The Xinjiang Uygur Handmade Carpets Pattern Vector Technology

Posted on:2014-02-22Degree:MasterType:Thesis
Country:ChinaCandidate:Y M AnFull Text:PDF
GTID:2248330398467384Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Xinjiang is the hometown of oriental rugs,handmade carpets of Xinjiang Uyguris not only well-known by domestic but also international reputation with itscharacteristics of fine workmanship,beautiful,brighter colors,fine texture etc..ButXinjiang Uygur handmade carpets has been serving to keep the hands of theworkshop-style production model,especially hand-painted carpet pattern iscomplicated and time-consuming process and require higher technial ofArtisan,hindering the efficiency of the production of carpets.This article researchedthe vector technology of Xinjiang Uygur handmade carpets for the purpose to impovethe drawn efficiency of carpet pattern and increase the diversity of patterns.Vector quantization is a hot research topic of computer graphics, patternrecognition, and the field of CAD/CAM, etc., at present the technology mainlyapplied in the vectorization of engineering drawings and maps. Vector quantizationfor the carpet pattern, especially for xinjiang carpet pattern which reflects nationalculture with national characteristics making it less researchers. Because the carpetpattern has the following characteristics: the patterns complex and changeable,thecurve is more complex, the patterns are susceptible to noise interference, pattern isdifficult to collect, etc..In this paper,a color pattern vector is proposed by analyzingthe characteristics of the Xinjiang Uygur carpet pattern and combining with theexisting research results.The process is divided into the following fivesteps:pre-processing,color clustering,contour extraction,key point detection,curvefitting.The article maked repeated test with the patterns of Qaxma carpet and provedthe feasibility of the proposed method.
Keywords/Search Tags:vector quantization, improved K-means clustering, edge detection, keypoint extraction, curve fitting
PDF Full Text Request
Related items