基于图像分解的织物组织自动识别

利用维纳滤波从水平和垂直两个方向分解织物图像,经过合理的二值化,分别获得包含经纱和纬纱特征信息的子图像,从而计算经纬纱密度等参数.综合子图像的部分信息,可以确定织物组织点的位置,再利用每个组织点区域内经纬向纹理特征,判断织物组织类型.经过对平纹、斜纹、缎纹3种基本组织布样的实验,表明该方法可以很好地识别织物组织类型.

Automatic Recognition of Fabric Structures Based on Image Decomposition

A method to recognize fabric structures automatically which decompose the fabric image into the vertical and horizontal directions subimages based on Wiener filter has been introduced.The parameters such as weaving density can be calculated from the thresholded subimages which include the warp and weft information respectively.The structure point areas can be defined by the texture information extracted from the subimages.The gray level difference in vertical direction and horizontal direction has been used to define the type of structure points.The result shows this method is feasible for plain,twill and stain patterns.