湖南侗锦图案的自动提取  

Automatic extraction of Hunan Dong brocade patterns

作　　者：王艺玮 郭虹[1] 易兵 WANG Yiwei;GUO Hong;YI Bing(College of Textile and Fashion,Hunan Institute of Engineering,Xiangtan 411104,China)

机构地区：[1]湖南工程学院纺织服装学院,湖南湘潭411104

出　　处：《丝绸》2023年第5期113-119,共7页Journal of Silk

基　　金：教育部人文社会科学研究项目(20YJE760001);湖南省社会科学成果评审委员会课题项目(XSP20YBZ101);湖南工程学院研究生科研创新项目(202127)。

摘　　要：为探寻民族织锦文化的最优传承模式,文章将湖南侗锦图案作为一种传统的文化符号与计算机图像处理方法相融合。由于侗锦图案在人工提取过程中存在直接取样难的问题,并且提取人员容易加入个人主观意愿,为了提高图案提取效率和准确度,对侗锦采取图像分析法进行自动提取。首先使用数码相机对侗锦图案进行有效采集,接着将图像灰度化,对处理后的图像通过中值滤波算法进行噪声平滑预处理,再利用大津阈值分割算法求取双阈值,最后使用Canny算法实现侗锦图案的自动提取。研究结果表明:该方法不仅能够提取出更多完整的边缘信息,且较人工手法而言更加精确、省时省力,较其他传统算法来说更有优势。Dong brocade,as one of the most beautiful brocades in China,gathers the religious belief and art aesthetics of the Dong people,making itself a treasure of China’s minority art civilization.Under the influence of multiple factors such as social modernization and economic integration,the Dong brocade culture is on the verge of disappearing and the protection situation is grim.With the rapid development of the digital era,the combination of Dong brocade patterns as a traditional cultural symbol and computer image processing can better adapt to the era of rapid development.Nowadays,the computer industry is widely used in the development and production of clothing,which provides a new technical approach for the extraction of traditional clothing patterns.In the paper,the research on the pattern extraction of Dong brocade in Hunan province was focused on the field investigation of the classic works of Dong brocade,and the pattern of Dong brocade was studied from the perspective of computer graphics and image processing.The automatic recognition and extraction of the patterns of Dong brocade by using image processing methods made the Dong brocade get better development and radiate new vitality.It is difficult to sample Dong brocade patterns directly in the process of manual extraction,and it is easy for the extractors to have their own subjective will.In order to improve the efficiency and accuracy of the pattern extraction,the automatic extraction of Dong brocade was carried out by the image analysis method.Firstly,Dong brocade patterns were collected effectively by digital camera,then the images were grayed,and the processed images were pre-processed by median filtering algorithm to smooth the noise,then the Otsu threshold segmentation algorithm was used to get the double thresholds,and finally the Canny algorithm was used to achieve the automatic extraction of Dong brocade patterns.To evaluate the edge detection effect of the proposed algorithm,the experimental segmentation results of the proposed algorithm were ev

关 键 词：传统文化 侗锦图案 图像处理 大津阈值分割算法 CANNY算法 

分 类 号：TS941.26[轻工技术与工程—服装设计与工程]