Hardware surface defect online detection method based on multi-modal sensing fusion
By employing a multimodal sensor fusion method, combining optical cameras, line laser scanners, and thermal imaging cameras, efficient and reliable detection of surface defects in hardware parts has been achieved. This addresses the limitations of single sensors in complex environments and improves detection accuracy and adaptability.
Patent Information
- Application Number
- CN202511592059.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-03
- Publication Date
- 2026-03-27
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing methods for detecting surface defects in hardware parts rely on a single sensor, which suffers from problems such as high reflectivity, interference from complex textures, low optical image contrast, and large variations in ambient lighting. This results in insufficient detection accuracy and robustness, and the lack of an adaptive mechanism makes it impossible to adapt to changes in different hardware parts types and production conditions.
A multimodal sensor fusion method is adopted, including a high-resolution optical camera, a line laser scanner and a thermal imaging camera. Color image data, 3D point cloud data and temperature distribution data are acquired through synchronous triggers. Reliability assessment and adaptive weighted fusion are performed by combining fuzzy logic method, and defect classification and localization are performed by support vector machine classifier.
It improves the accuracy and robustness of surface defect detection for hardware parts, can efficiently identify minute defects under different lighting conditions, adapts to complex production environments, reduces false detections and missed detections, and improves the overall efficiency and adaptability of the production line.