System and method for defect detection based on synthetic defect images
MYPI2024007733A0Pending Publication Date: 2026-06-30JARING COMMUNICATIONS SDN BHD
0 Cites 0 Cited by
Patent Information
- Application Number
- MYPI2024007733
- Authority / Receiving Office
- MY · MY
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-12-31
- Publication Date
- 2026-06-30
Abstract
The present invention relates to a system (100) for defect detection based on synthetic defect images. The system (100) comprises a data preparation module (102) for collecting and annotating images, a synthetic data generation module (104) for generating synthetic defect images, and a class incremental training module (108) for training a defect detection model using real-world and synthetic images. The defect detection model is subsequently used to perform defect detection on real-world images. Additionally, a model generation module (110) optimizes the defect detection model, and an object identification module (112) visualizes identified defects. The present invention also describes a method (900) for defect detection based on the synthetic defect images. The method involves receiving images from a data capturing module, generating synthetic defect images, evaluating their accuracy, and training the defect detection model with real-world images and synthetic defect images. The most illustrative drawing is Figure 1
Need to check novelty before this filing date? Find Prior Art
