Knowledge Commons of Institute of Automation,CAS
EDDs: A series of Efficient Defect Detectors for fabric quality inspection | |
Zhou, Tong1,3; Zhang, Jiabin2,3![]() ![]() ![]() | |
发表期刊 | MEASUREMENT
![]() |
ISSN | 0263-2241 |
2021-02-01 | |
卷号 | 172期号:1页码:8 |
摘要 | Deep Convolutional Neural Network (DCNN) has recently advanced state-of-the-art performance on vision-related tasks and its application is further extended to industrial fields. The paper focuses on the problem of fabric defect detection to which an efficient DCNN architecture is developed. In contrast to previous methods that directly apply existing DCNN models demonstrated on natural images to industrial images, the proposed Efficient Defect Detectors (EDDs) are sufficiently optimized with consideration of the characteristics of fabric surface images, i.e., resolution, defect appearance, etc. Firstly, lightweight backbone is suggested in EDD to improve computational efficiency without reduction in image resolution. Secondly, a new feature fusion strategy named L-shaped feature pyramid network (L-FPN) is proposed and utilized to make full use of low-level texture features which are demonstrated to be more important than high-level semantic features in defect recognition. Based on the configurations of lightweight backbone and L-FPN, we use only one hyper-parameter to jointly adjust the proportion of resources occupied by width, depth and input resolution so that a family of defect detectors under different resource constraints can be developed. Experiments are conducted on a large fabric dataset to demonstrated the effectiveness of EDDs. Compared with the recent state-of-the-art detector, EfficientDet-d3, EDD-d3 achieves higher mean Average Precision (mAP) (20.9 vs 19.9) but with fewer parameters. EDD-d3 has 8.59M parameters and 31.78B FLOPs (floating point operation per second), which respectively are 39.8% and 49.0% lower than EfficientDet-d3. The proposed EDDs achieve better tradeoff between accuracy and speed than previous methods. EDDs could be applied to fabric production sits with different resource restrictions, which demonstrates that EDDs have important application value. |
关键词 | Defect detection Convolutional neural network Fabric quality inspection Feature fusion |
DOI | 10.1016/j.measurement.2020.108885 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | National Key Research and Development Program of China[2018YFB1306303] ; National Natural Science Foundation of China[61773374] ; National Natural Science Foundation of China[61702323] ; Major Basic Research Projects of Natural Science Foundation of Shandong Province, China[ZR2019ZD07] |
项目资助者 | National Key Research and Development Program of China ; National Natural Science Foundation of China ; Major Basic Research Projects of Natural Science Foundation of Shandong Province, China |
WOS研究方向 | Engineering ; Instruments & Instrumentation |
WOS类目 | Engineering, Multidisciplinary ; Instruments & Instrumentation |
WOS记录号 | WOS:000619231500004 |
出版者 | ELSEVIER SCI LTD |
七大方向——子方向分类 | 目标检测、跟踪与识别 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/43995 |
专题 | 中科院工业视觉智能装备工程实验室_精密感知与控制 |
通讯作者 | Su, Hu |
作者单位 | 1.Chinese Acad Sci, Inst High Energy Phys, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Zhou, Tong,Zhang, Jiabin,Su, Hu,et al. EDDs: A series of Efficient Defect Detectors for fabric quality inspection[J]. MEASUREMENT,2021,172(1):8. |
APA | Zhou, Tong,Zhang, Jiabin,Su, Hu,Zou, Wei,&Zhang, Bohao.(2021).EDDs: A series of Efficient Defect Detectors for fabric quality inspection.MEASUREMENT,172(1),8. |
MLA | Zhou, Tong,et al."EDDs: A series of Efficient Defect Detectors for fabric quality inspection".MEASUREMENT 172.1(2021):8. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
EDDs.pdf(1132KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论