Research on the Classification of Cashmere and Wool Fibers, Based on SVMs

摘要：

Cashmere and wool fibers are highly valuable industrial products. It is necessary to differentiate them. In this paper, 6 types of cashmere and wool fibers samples are obtained from different places (more 200 samples of each type). The digital images are captured by CCD camera under Axioskop 2 MOT PLUS differences contrast (DIC) microscope. Each fiber has seven features extracted by manual work. The support vector machines (Sims) are used to classify cashmere and wool. The test result shows that the minimum classification error rate is 4.33％ when the total number of training samples is 510 and sigma=9.

关键词： SVMs cashmere wool classification identification

作者: Yaxia Liu Kan Shi Shuyuan Shang

作者单位: Department of Information Engineering, Beijing Institute of Fashion Technology, Beijing, China

会议类型: 国际会议

会议名称: 2010 International Conference on Image Analysis and Signal Processing(2010 图像分析与信号处理国际会议 IASP 10)

会议地点: 厦门

会议语种:英文

页码: 186-189

在线出版日期: 2010-04-12（万方平台首次上网日期，不代表论文的发表时间）