Connector quality detection method and system based on visual measurement
By combining multi-view image acquisition and deep neural network recognition technology with adaptive illumination compensation and noise reduction processing, the problems of weak anti-interference ability and low accuracy of micron-level defect identification in connector quality inspection have been solved, realizing high-precision connector quality inspection with low false detection rate on high-speed production lines.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN FJY ELECTRONICS
- Filing Date
- 2026-04-25
- Publication Date
- 2026-07-17
AI Technical Summary
Existing connector quality inspection technologies have weak anti-interference capabilities on high-speed production lines, low accuracy in identifying micron-level minute defects, and the inability to adaptively optimize inspection rules. They also lack a linkage and traceability mechanism between defects and production processes, resulting in high false detection and false negative rates, and failing to meet the requirements for high-precision and high-stability online quality inspection.
By employing multi-view image acquisition combined with adaptive illumination compensation and noise reduction preprocessing, and using a lightweight deep neural network for multi-scale feature fusion recognition, a defect end-to-end traceability mechanism is established, detection parameters are dynamically adapted, and production data feedback optimization is achieved.
It effectively reduces the risk of false detection and missed detection, accurately identifies micron-level defects, improves detection accuracy and real-time performance, adapts to the needs of high-speed production lines, realizes adaptive optimization of detection parameters and full-chain traceability of defects, and meets the requirements of high-precision and high-stability online quality inspection.
Smart Images
Figure CN122415548A_ABST