wear debris

wear debris classification with quantity and size calculation using convolutional neural networks

The majority of faults in steel production equipment are attributed to wear and tear. Monitoring the wear status of these machines can be effectively achieved by classifying wear debris in their lubrication systems. In this project, we utilize Convolutional Neural Networks (CNNs) to classify wear debris. Additionally, we employ the Faster R-CNN model to detect the quantity and size of the wear debris. To enhance our results further, we integrate the GrabCut algorithm, which aids in refining the accuracy of our detection and classification processes.

will update soon…

References

2019

  1. Wear debris classification and quantity and size calculation using convolutional neural network
    Hongbing Wang, Fei Yuan, Liyuan Gao, Rong Huang, and Weishen Wang
    In Cyberspace Data and Intelligence, and Cyber-Living, Syndrome, and Health: International 2019 Cyberspace Congress, CyberDI and CyberLife, Beijing, China, December 16–18, 2019, Proceedings, Part I 3, 2019

2018

  1. Wear debris classification of steel production equipment using feature fusion and case-based reasoning
    Hongbing Wang, Rong Huang, Liyuan Gao, Weishen Wang, Anjun Xu, and 1 more author
    ISIJ International, 2018