Particleboard Surface Defect Inspection Based on Data Augmentation and Attention Mechanisms

Baizhen Li, Zhijie Xu, En Kai Bian, Chen Yu, Feng Gao, Yanlong Cao

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Inspection accuracy of surface defect is very important for particleboard production. However, the insufficient defect samples seriously restrict the quality of vision and deep learning-based inspection result. The small-scale defects on particleboard surface are also a major challenge to the input of network models. This paper proposes a method based on data augmentation and attention mechanisms to solve these problems. A hardware platform was designed to take surface defect images. The methods of traditional data augmentation and GAN have been applied to increase the amount of defect samples. The Poisson Fusion technique was adopted to generate defect images albeit varied backgrounds to for network training. The SSD network was deployed as the optimization model. The devised optimization schemes replaced the feature extraction network (VGG) with ResNET18 and ResNET50 respectively before fusing with the DCGAN module. During the training stage, a transfer learning-based method was developed to pre-train the optimized network through COCO2017 dataset to improve the training speed and accuracy. The experimental results showed that the scheme of "ResNET50 + Attention"outperformed benchmarked solutions with a peak performance on particleboard surface defect inspection reaching 96.79%.

Original languageEnglish
Title of host publication2022 27th International Conference on Automation and Computing
Subtitle of host publicationSmart Systems and Manufacturing, ICAC 2022
EditorsChenguang Yang, Yuchun Xu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages6
ISBN (Electronic)9781665498074
ISBN (Print)9781665498081
DOIs
Publication statusPublished - 10 Oct 2022
Event27th International Conference on Automation and Computing - Bristol, United Kingdom
Duration: 1 Sep 20223 Sep 2022
Conference number: 27

Conference

Conference27th International Conference on Automation and Computing
Abbreviated titleICAC 2022
Country/TerritoryUnited Kingdom
CityBristol
Period1/09/223/09/22

Fingerprint

Dive into the research topics of 'Particleboard Surface Defect Inspection Based on Data Augmentation and Attention Mechanisms'. Together they form a unique fingerprint.

Cite this