论文信息 - Data Augmentation Method For Fabric Defect Detection

Data Augmentation Method For Fabric Defect Detection

This paper proposes a multi-channel faster-RCNN for fabric defect detection. Since the fabric defects are not trivial, this paper uses geometric data augmentation and GAN-based data augmentation to increase the training samples. The proposed method significantly improves the conventional faster-RCNN's performance in the simulation. The average accuracy of the proposed multi-channel faster-RCNN with an augmented dataset is up to 90.05%.

Chien-Chou Lin | Po-Hsiang Wang

[1] Steven Tien Guan Thng,et al. Classification of Non-Tumorous Facial Pigmentation Disorders Using Improved Smote and Transfer Learning , 2019, 2019 IEEE International Conference on Image Processing (ICIP).

[2] Wai Keung Wong,et al. Differential evolution-based optimal Gabor filter model for fabric inspection , 2016, Neurocomputing.

[3] Kaiming He,et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4] Yoshua Bengio,et al. Generative Adversarial Nets , 2014, NIPS.

[5] Qian Yang,et al. The fabric defect detection technology based on wavelet transform and neural network convergence , 2013, 2013 IEEE International Conference on Information and Automation (ICIA).

[6] Sun Zhi-gang,et al. Textural Fabric Defect Detection using Adaptive Quantized Gray-level Co-occurrence Matrix and Support Vector Description Data , 2012 .

[7] Thanos Stouraitis,et al. Defect detection and classification on web textile fabric using multiresolution decomposition and neural networks , 1999, ICECS'99. Proceedings of ICECS '99. 6th IEEE International Conference on Electronics, Circuits and Systems (Cat. No.99EX357).

[8] George Papadopoulos,et al. Real-time vision system for defect detection and neural classification of web textile fabric , 1999, Electronic Imaging.

[9] Frederick Y. Wu,et al. Machine Vision Applications in Industrial Inspection , 1993 .