摘要
为快速准确地鉴别羊绒和羊毛,提出一种基于视觉词袋模型的鉴别方法。该方法使用羊绒和羊毛的光学显微镜图像作为实验样本,将纤维鉴别问题转化为图像的分类问题。首先对光学显微镜图像进行预处理以增强特征,然后从纤维形态中提取局部特征并生成视觉单词,再依据视觉单词对纤维图像进行分类,从而达到鉴别纤维的目的。使用了4 400幅纤维图像作为数据集,从中选择不同的羊绒和羊毛的混合比作为训练集和测试集,得到的识别率最高为86%,最低为81.5%,鉴别1 000根纤维需要的时间小于100 s,训练好的分类器可保存并用于后期的检测工作。
In order to identify cashmere and wool rapidly and accurately,a method based on bag-ofvisual-word was proposed. Optical microscope images of cashmere and wool were taken as experimental samples in this method. The problem of fiber identification was changed to the problem of image classification. Firstly,fiber images were pre-processed to enhance their characteristics. Then,local features were extracted from fiber morphology and these local features were converted to visual words.Fiber images can be classified using visual words mentioned above. The experimental dataset contains4 400 fiber images. Different mixing ratio of cashmere and wool were selected as train set and test set from the dataset. In this experiment,the highest recognition ratio is 86%,and the lowest is 81. 5%. The time required to identify 1 000 fibers is shorter than 100 s. The trained classifier can be saved and used for the late detection.
出处
《纺织学报》
EI
CAS
CSCD
北大核心
2017年第7期130-134,141,共6页
Journal of Textile Research
基金
国家自然科学基金资助项目(61572124)
中央高校基本科研业务费专项资金资助项目(CUSF-DH-D-2016016)
上海市自然科学基金资助项目(14ZR1401100)
关键词
羊绒
羊毛
视觉词袋模型
图像处理
快速鉴别
cashmere
wool
bag-of-visual-word
image processing
rapid identification
作者简介
路凯（1979-），男，讲师，博士生。研究方向为纤维图像鉴别、纺织图像技术。
钟跃崎，通信作者，E-mail：zhyq@dhu．edu．cn。
