LED light splitting ribbon real-time defect detection compensation method and system based on edge computing
By using a lightweight neural network model and real-time image processing on an edge computing device, the problems of real-time performance, accuracy, and network dependence in LED beam splitting and tape-making inspection were solved. High-precision defect detection and position compensation under high-speed motion were achieved, improving detection accuracy and production stability, and supporting continuous iteration of models and processes.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN SITUOAN OPTOELECTRONICS CO LTD
- Filing Date
- 2026-04-22
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for LED beam splitting and tape-making inspection suffer from a contradiction between real-time performance and accuracy, strong dependence on networks, difficulty in model updates, and low data utilization, making it difficult to achieve high-precision defect detection and micron-level position compensation under high-speed motion.
A real-time defect detection method based on edge computing is adopted. A lightweight deep neural network model is used to perform image processing on an edge computing device. Combined with motion blur restoration and adaptive threshold segmentation, real-time rejection and compensation instructions are generated and uploaded to the cloud system asynchronously to build a closed loop of edge AI inference and control.
It achieves a detection-compensation response within 50ms, with a position compensation accuracy of ±10 micrometers and a detection accuracy of 99.5%, reducing network dependency risks and enabling data-driven continuous optimization.
Smart Images

Figure CN122415522A_ABST