A dual-branch fabric defect detection method, storage medium and electronic device
Through the dual-branch object detection neural network model and image stitching technology, the accuracy and environmental sensitivity of grey fabric textile defect detection are solved, and high-precision defect recognition is achieved.
CN115358967BActive Publication Date: 2025-08-26SHANGHAI ZHIJING INFORMATION TECH CO LTD
Patent Information
- Application Number
- CN202210816951.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-12
- Publication Date
- 2025-08-26
- Estimated Expiration
- 2042-07-12
Smart Images
Figure CN115358967B_ABST
Abstract
The present invention discloses a dual-branch fabric defect detection method. The present invention adopts a dual-network architecture method for network architecture, which can effectively improve the accuracy of detection results; uses a deep learning neural network to detect grey cloth defects based on image content comparison, which can greatly improve the feature representation of subtle defect detection. The extraction of this feature representation incorporates data of good products, and the deep learning network has richer data information. In addition, the image size is transformed using an image pyramid, which can extract defects of smaller size.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Hierarchical image classification system
US20140270347A1
Tongue image extraction method and device, and a computer readable storage medium
WO2020215697A1