基于图像处理技术的织物组织识别研究现状  

Research Status of Fabric Weave Identification Based on Image Processing Technology

作　　者：张瑞[1] 辛斌杰[1] 

机构地区：[1]上海工程技术大学,上海201620

出　　处：《棉纺织技术》2015年第11期80-84,共5页Cotton Textile Technology

基　　金：国家自然科学基金(61271419)

摘　　要：总结基于图像处理技术的织物组织识别研究现状。在织物组织识别过程中,特征参数的提取和织物组织的识别是关键问题。针对这两个问题,概括和分析了近年来国内外研究人员的研究成果,包括基于频率域的傅立叶变换和小波变换、基于空间域的灰度共生矩阵、自相关函数、基于神经网络的方法和基于聚类分析的方法等;总结了当前织物识别研究中存在的不足。认为:基于图像处理技术的织物组织识别方法是建立在客观评价基础上的,具有准确、快速、效率高等优势,在纺织品检测领域具有良好的发展前景。The research status of fabric weave identification based on image processing technology was sum- marized. During the identification of fabric weave, the extracting of characteristic parameters and the identification of fabric weave were the key problems. Focusing on the two problems, research achievement of researchers at home and abroad in recent years were summarized and analyzed including Fourier transform and wavelet transform based on frequency domain, gray-level co-occurrence matrix based on spatial domain, autocorrelation function, methods based on neural network, methods based on clustering analysis and so on. The disadvantages existed in current fabric identification research were summarized. It is considered that the fabric weave identification method based on image processing technology is established under the foundation of objective evaluation. It has advantages including accuracy, fast, and higher efficiency and so on. It has greater developing prospect in textile testing fields.

关 键 词：机织物 织物组织 特征参数 傅立叶变换 小波变换 灰度共生矩阵 神经网络 

分 类 号：TS107[轻工技术与工程—纺织工程]