首页> 中文期刊>武汉纺织大学学报 >贾卡经编织物花纹图案的自动识别

贾卡经编织物花纹图案的自动识别

     

摘要

在一般的花型设计和检测印染织物花型是否走样的过程中,都需要将花纹提取出来。为了快速地将织物花纹图案提取出来,提高花纹设计效率,本文主要基于织物图像的颜色特征,对织物图像建立高斯混合模型,然后根据后验概率的排序大小,并对织物图像的不同层次进行相应的标注,从而标记出所需的织物花纹。结果表明:该方法能有效地将完整的花纹提取出来,准确度高,速度快,具有一定的实用价值。%Generally, in the process of designing from samples from and testing pattern whether its shape is different or not all, we need to extract the pattern. In order to quickly and efficiently extract the fabric patterns automatically and improve the efficiency of decorative pattern design, the method is mainly based on color feature, to establish the Gaussian mixture model, according to the posteriori probability, and then image annotation. Results show that the method can effectively extract the complete pattern, with high accuracy, high speed, and has a certain practical value.

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号