Patent · US Active

Neural network-based defect detection method for gluing quality on aircraft skin

US12423796B1 · kind B1 · utility

0Cited by
0References
7Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJun 9, 2025
Grant dateSep 23, 2025
Priority date
Expiry dateJun 9, 2045

Classification

  • Technology area (CPC Y)Emerging Cross-Sectional Technologies
  • CPC primaryY02P90/30
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Disclosed in the present invention is a neural network-based defect detection method for gluing quality on aircraft skin. The method includes: data acquisition: taking photos of aircraft skin by using a camera to acquire image data; preprocessing the acquired image data; annotating the data by using annotation software to acquire a data set for network training; establishing a defect detection network model based on feature erasure and boundary refinement, where the defect detection network model includes a feature extraction network, a semantic-guided feature erasure module, a multi-scale feature fusion network, and a defect prediction network based on boundary refinement, which are sequentially connected, the data set is used for training the network model, and trained model parameters are saved; and detecting a directly collected skin gluing image by using the trained network model and outputting detection results.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.