Skip to main content
placeholder image

Defect detection and classification of galvanized stamping parts based on fully convolution neural network

Conference Paper


Abstract


  • In this paper, a new convolution neural network method is proposed for the inspection and classification of galvanized stamping parts. Firstly, all workpieces are divided into normal and defective by image processing, and then the defective workpieces extracted from the region of interest (ROI) area are input to the trained fully convolutional networks (FCN). The network utilizes an end-to-end and pixel-to-pixel training convolution network that is currently the most advanced technology in semantic segmentation, predicts result of each pixel. Secondly, we mark the different pixel values of the workpiece, defect and background for the training image, and use the pixel value and the number of pixels to realize the recognition of the defects of the output picture. Finally, the defect area's threshold depended on the needs of the project is set to achieve the specific classification of the workpiece. The experiment results show that the proposed method can successfully achieve defect detection and classification of galvanized stamping parts under ordinary camera and illumination conditions, and its accuracy can reach 99.6%. Moreover, it overcomes the problem of complex image preprocessing and difficult feature extraction and performs better adaptability.

Authors


  •   Xiao, Zhitao (external author)
  •   Leng, Yanyi (external author)
  •   Geng, Lei (external author)
  •   Xi, Jiangtao

Publication Date


  • 2018

Citation


  • Z. Xiao, Y. Leng, L. Geng & J. Xi, "Defect detection and classification of galvanized stamping parts based on fully convolution neural network," in 9th International Conference on Graphic and Image Processing (ICGIP 2017), 2018, pp. 1-8.

Scopus Eid


  • 2-s2.0-85046453133

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers1/1342

Start Page


  • 1

End Page


  • 8

Place Of Publication


  • Bellingham, United States

Abstract


  • In this paper, a new convolution neural network method is proposed for the inspection and classification of galvanized stamping parts. Firstly, all workpieces are divided into normal and defective by image processing, and then the defective workpieces extracted from the region of interest (ROI) area are input to the trained fully convolutional networks (FCN). The network utilizes an end-to-end and pixel-to-pixel training convolution network that is currently the most advanced technology in semantic segmentation, predicts result of each pixel. Secondly, we mark the different pixel values of the workpiece, defect and background for the training image, and use the pixel value and the number of pixels to realize the recognition of the defects of the output picture. Finally, the defect area's threshold depended on the needs of the project is set to achieve the specific classification of the workpiece. The experiment results show that the proposed method can successfully achieve defect detection and classification of galvanized stamping parts under ordinary camera and illumination conditions, and its accuracy can reach 99.6%. Moreover, it overcomes the problem of complex image preprocessing and difficult feature extraction and performs better adaptability.

Authors


  •   Xiao, Zhitao (external author)
  •   Leng, Yanyi (external author)
  •   Geng, Lei (external author)
  •   Xi, Jiangtao

Publication Date


  • 2018

Citation


  • Z. Xiao, Y. Leng, L. Geng & J. Xi, "Defect detection and classification of galvanized stamping parts based on fully convolution neural network," in 9th International Conference on Graphic and Image Processing (ICGIP 2017), 2018, pp. 1-8.

Scopus Eid


  • 2-s2.0-85046453133

Ro Metadata Url


  • http://ro.uow.edu.au/eispapers1/1342

Start Page


  • 1

End Page


  • 8

Place Of Publication


  • Bellingham, United States