欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为了对具有纹理图案的纺织品缺陷进行准确有效的检测,提出了一种新的基于规则带（regular band,RB）特征的纹理图案布匹瑕疵检测方法。首先利用规则带分析提取图像的各子带纹理特征;然后对各子带图像特征求梯度,并通过融合各子带梯度来获得纹理梯度,使其在纹理梯度中能有效地突出纹理区域的边界;最后在此基础上,结合分水岭分割,准确地检测出纺织品的缺陷区域。实验结果表明,新算法能够准确地定位图案纹理中的瑕疵区域,且操作简便,鲁棒性强。
英文摘要：
A novel method for the Patterned Texture Inspection based on regular band is proposed.The properties of texture defects are characterized using the regularity analysis,then the texture gradient from every sub-band is calculated.Texture gradient contains many texture features and the boundary of texture regions in patterned images can be enhanced.Combined with the watershed segmentation,the defective regions in patterned images can be detected accurately.Experimental results show that the method can detect defective regions accurately,and it is easy to operate and robust.
同期刊论文项目
同项目期刊论文
期刊信息
位置：