文献类型：期刊文献

中文题名：图像分割算法在花型自动识别中的应用

英文题名：Application of an image segmentation algorithm in pattern auto-recognition

作者：万永菁[1];万光逵[2]

机构：[1]华东理工大学信息科学与工程学院,上海200237;[2]南昌航空工业学院电子工程系,江西南昌330034

年份：2007

卷号：28

期号：5

起止页码：63

中文期刊名：纺织学报

外文期刊名：Journal of Textile Research

收录：CSTPCD;;Scopus;北大核心:【北大核心2004】；CSCD:【CSCD2011_2012】；

语种：中文

中文关键词：模糊C均值聚类(FCM);图像分割;空间信息;花型识别

外文关键词：fuzzy C-means clustering; image segmentation; spatial information; pattern recognition

摘要：针对提花毛皮样片扫描图像的花型自动识别问题,提出了一种改进的基于空间信息的模糊C均值聚类图像分割算法,该算法利用像点邻域区间的粗糙度和邻域像点的值修正像点与聚类中心的距离,对像点的模糊隶属度函数进行修正,利用修正后的模糊隶属度函数进行聚类中心的迭代计算,获得合理的聚类中心。经多幅提花毛皮样片的花型图像分割实验表明,该算法具有对噪声不敏感的优点,在进行提花毛皮样片扫描图像的花型识别时,能获得较好识别结果。
Aiming at the problem of pattern auto-recognition of jacquard fur sample scanning image, an improved fuzzy C-means clustering image segmentation algorithm based on spatial information is proposed, The distance between the pixel and the clustering center is modified by using the coarseness degree of the neighbor field and the value of the pixel, and the fuzzy membership function of the pixel is modified. The reasonable clustering center is gained by iterative calculation using modified fuzzy membership function. Mter several experiments of jacquard fur sample scanning image segmentation, it is indicated that the algorithm has the advantage of less sensitive to noise and it can get preferable recognition results when recognizing the pattern of jacquard fur sample scanning image.