Improvement of High-Speed Detection Algorithm for Nonwoven Material Defects Based on Machine Vision

Abstract

Defect detection is vital in the nonwoven material industry, ensuring surface quality before producing finished products. Recently, deep learning and computer vision advancements have revolutionized defect detection, making it a widely adopted approach in various industrial fields.This paper mainly studied the defect detection method for nonwoven materials based on the improved Nano Det-Plus model. Using the constructed samples of defects in nonwoven materials as the research objects, transfer learning experiments were conducted based on the Nano DetPlus object detection framework. Within this framework,the Backbone, path aggregation feature pyramid network(PAFPN)and Head network models were compared and trained through a process of freezing, with the ultimate aim of bolstering the model's feature extraction abilities and elevating detection accuracy. The half-precision quantization method was used to optimize the model after transfer learning experiments, reducing model weights and computational complexity to improve the detection speed.Performance comparisons were conducted between the improved model and the original Nano Det-Plus model,YOLO, SSD and other common industrial defect detection algorithms, validating that the improved methods based on transfer learning and semi-precision quantization enabled the model to meet the practical requirements of industrial production.

Keywords

defect detection / nonwoven materials / deep learning / object detection algorithm / transfer learning / half-precision quantization

References

[1] ZHANG R T, Qi X N. A study of China’s nonwovens industry based on the SCP paradigm[J]. Information Systems and Economics, 2022, 3(4): 1-8.

[2] LIU S P. Research on the mechanism and path of the manufacturing industry upgrading with “ Internet Plus ”[D]. Changsha: Zhongnan University of Economics and Law, 2019. (in Chinese)

[3] XU W F, DU J. Discussion on green innovation and development of textile industry enabled by digital technology[J]. Cotton Textile Technology, 2023, 51(8): 33-37.(in Chinese)

[4] NGAN H Y T, PANG G K H, YUNG N H C. Automated fabric defect detection:a review[J]. Image and Vision Computing, 2011, 29(7): 442-458.

[5] BOU NASSIF A, ABU TALIB M, NASIR Q, et al. Machine learning for anomaly detection: a systematic review[J]. IEEE Access, 2021, 9: 78658-78700.

[6] ZHU R H, XIN B J, DENG N, et al. Review of fabric defect detection based on computer vision[J]. Journal of Donghua University(English Edition), 2023, 40(1): 18-26.

[7] LI F, PAN H S, SHENG S X, et al. Image retrieval based on vision transformer and masked learning[J]. Journal of Donghua University(English Edition), 2023, 40(5): 539-547.

[8] KHAN A, LAGHARI A, AWAN S. Machine learning in computer vision: a review[J]. ICST Transactions on Scalable Information Systems, 2021, 8(32): 169418.

[9] LI M, YANG S, HE R H, et al. Jacquard fabric defect detection technology combining contextawareness and convolutional neural network[J]. Modern Textile Technology, 2021, 29(6): 62-66. (in Chinese)

[10] WEI W, DENG D X, ZENG L, et al. Real-time implementation of fabric defect detection based on variational automatic encoder with structure similarity[J]. Journal of Real-Time Image Processing, 2021, 18(3): 807-823.

[11] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. New York: ACM, 2014: 580-587.

[12] ZHANG W Y, YANG Y, LIN C, et al. Method of fabric defect detection based on Gabor filter and BP neural network[J]. Journal of Zhongyuan University of Technology, 2014, 25(3): 1-6. (in Chinese)

[13] LI Y D, ZHANG J Y, LIN Y B. Combining fisher criterion and deep learning for patterned fabric defect inspection[J]. IEICE Transactions on Information and Systems, 2016, 99(11): 2840-2842.

[14] JING J F, FAN X T, LI P F, et al. Yarn-dyed fabric defect detection based on deepconvolutional neural network[J]. Journal of Textile Research, 2017, 38(2): 68-74.(in Chinese)

[15] LIU F J, LI Y. SAR remote sensing image ship detection method NanoDet based on visual saliency[J]. Journal of Radars, 2021, 10(6): 885-894. (in Chinese)

[16] XU Y T. Defect detection of automotive composite leather based on NanoDet-Plus[D]. Wuhan: Jianghan University, 2023. (in Chinese)

[17] QI P Y, WANG H Y, ZHANG J, et al. Crowded pedestrian detection algorithm based on improved FCOS[J]. CAAI Transactions on Intelligent Systems, 2021, 16(4): 811-818.

[18] REZATOFIGHI H, TSOI N, GWAK J, et al. Generalized intersection over union: a metric and a loss for bounding box regression[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Los Alamitos, IEEE, 2019: 658-666.

[19] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[EB/OL].( 2020-10-23)[2024-4-15]. https://arxiv.org/abs/2004.10934v1.

[20] HUGHES D P, SALATHE M. An open access repository of images on plant health to enable the development of mobile disease diagnostics[EB/OL]. (2015-11-25)[2024-4-15]. https://arxiv.org/abs/1511.08060.

[21] LI X L, LI Z X, YANG X F, et al. Boosting the generalization ability of Vis-NIR-spectroscopybased regression models through dimension reduction and transfer learning[J]. Computers and Electronics in Agriculture, 2021, 186: 106157.

[22] PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359.

Funding

National Key Research and Development Program of China(2022YFB4700600)

National Key Research and Development Program of China(2022YFB4700605)

National Natural Science Foundation of China(61771123)

National Natural Science Foundation of China(62171116)

Fundamental Research Funds for the Central Universities and Graduate Student Innovation Fund of Donghua University, China(CUSF-DH-D-2022044)