Improving the efficiency of steel plate surface defect classification by reducing the labelling cost using deep active learning

HIGHLIGHTS

• What: These methods only utilize the model predictions on the unlabeled data, ignoring the uncertainty information of the model on the labeled data, which is considered useful. In this work, a deep active learning method for steel plate defect classification is proposed. The model is trained with DL; 11 DU is used to verify the performance of the model and VA` is calculated; 12 Until VA` > VA. The experiment is mainly carried out in two parts.
• Who: Yang and colleagues from the Xiangtan University, School of Mechanical Engineering and Mechanics, China have published the research . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.