基于贝叶斯方法的织物分类研究

Textile Recognition Based on Bayesian Approach

提出基于贝叶斯决策理论的织物图像自动分类的新方法。先对织物图像的分类问题建立贝叶斯模型 ,再提取织物图像的形态结构参数作为特征向量 ,并计算出分类结果。该法能准确、快速、有效地对各类织物实现自动分类识别。

A new method automatically identifying textile is proposed in this paper.Firstly,Bayesian Model is presented to solve the problem of textile image classification.Secondly,the shape and structure parameters of the textile image are extracted as the feature vectors.Finally,the classification result is computed by the Bayesian formula.It can be seen from the experiment that the textile can be identified and classified quickly and exactly,it is proved that the method adopted is very feasible.