高级检索

彩色图像的织物组织自动识别

Automatic Identifying Fabric Structure Based on Color Images

  • 摘要: 针对基于彩色图像的单层组织识别问题,将二重织物的组织识别问题分解为单组织识别与整合.为了提高单层组织自动识别的精度,运用颜色聚类等方法分割织物样图,并提出了一种经纱分割算法,实现了经纬纱线的准确分割.最后,提出了一个基于织物表面图像的半自动交互式纬二重组织识别方案.实验结果表明该算法是有效的.

     

    Abstract: We identify the single-layer fabric structure given as a color image, then identification of the double-layer fabric is regarded as the combination of several single-layer fabrics. Color clustering is used to segment the fabric image, and a new algorithm of warp yarn segmentation is proposed to perform the identification more precisely. Finally, with human-computer interaction, a method of semi-automatic identification of double-layer fabric based on the surface images of the fabric is proposed. Experimental outcome shows availability of the method.

     

/

返回文章
返回