Automatic Classification of Fabric Flatness Templates Based on Deep Learning

摘要：

　　The automatic classification of fabric flatness levels has a great influence on the garment industry.The extraction of features based on artificial or neural networks has problems such as time-consuming and low accuracy.For this reason, deep convolutional neural network models are used to deeply learn the data of fabric templates.In this paper, the gray value and height of the fabric template are selected as the input of the model, and a convolutional neural network model is constructed to train the data.The experimental results show that the height value is easier to extract features, the classification accuracy can reach 98%, and it has good robustness and generalization.

关键词： Flatness rating Deep learning Convolutional neural network Template height

作者: Wenjun Zhang Zhu Zhan Jun Wang

作者单位: College of Textiles, Donghua University, Shanghai 201620, China Key Lab of Textile Science and Technology, Ministry of Education, Shanghai 201620, China College of Textiles, Donghua University, Shanghai 201620, China;Key Lab of Textile Science and Techn

会议类型: 国际会议

会议名称: 2018 Sino-Africa International Symposium on Textiles and Apparel (SAISTA2018)(2018中非纺织服装国际论坛)

会议地点: 巴赫达尔

会议语种:英文

页码: 379-382

在线出版日期: 2018-09-15（万方平台首次上网日期，不代表论文的发表时间）