Glass surface defect detection method

WO2026016074A1PCT designated stage Publication Date: 2026-01-22SHANGHAI UNIV OF MEDICINE & HEALTH SCI
3 Cites -1 Cited by

Patent Information

Application Number
PCT/CN2024/105873
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2026-01-22

Smart Images

  • Figure CN2024105873_22012026_PF_FP_ABST
    Figure CN2024105873_22012026_PF_FP_ABST
Patent Text Reader

Abstract

The present application discloses a glass surface defect detection method, relating to the field of computer vision. The method comprises: acquiring a glass defect image to undergo detection; and preprocessing the glass defect image to undergo detection, so as to obtain a multi-resolution glass defect image. In the method, a multi-scale convolutional neural network model is designed by incorporating image Gaussian difference pyramid decomposition into a conventional convolutional neural network. The method further comprises: on the basis of the multi-resolution glass defect image, determining a defect category by using a designed glass surface defect detection model. The glass surface defect detection model is obtained by using a training data set to train the multi-scale convolutional neural network model. The structure of the multi-scale convolutional neural network model comprises an input layer, three convolutional layers, three pooling layers, three fully connected layers, and a Softmax classifier. The present application improves the accuracy of glass surface defect detection.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Convolutional neural network-based photovoltaic glass defect classification method and device

    CN108108768A

  • Glass defect detection method and device based on deformable convolution and feature fusion

    CN114170230A

  • Determining image defects using image comparisons

    US20220036525A1