一种基于计算机视觉的线路板铜箔厚度检测方法
By combining multispectral imaging and machine learning models, adaptive detection of copper foil thickness on circuit boards was achieved, solving the thickness detection problem in complex scenarios with multiple materials, improving detection accuracy and stability, reducing reliance on manual calibration, and demonstrating strong adaptability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- LONGYU ELECTRONICS MEIZHOU
- Filing Date
- 2025-09-09
- Publication Date
- 2026-07-17
AI Technical Summary
Existing copper foil thickness detection technologies for circuit boards are ill-suited to complex scenarios involving multiple materials and interfaces. Drastic changes in material properties lead to significant fluctuations in thickness calculation results, and the lack of dynamic model iteration and real-time correction capabilities makes it difficult to meet the requirements for high precision and automatic dynamic self-adaptation.
Multispectral imaging is used to acquire multispectral pixel data of different blocks of the circuit board. Spectral reflectance curves and surface texture features are extracted through normalization and reflectance standardization. Thickness prediction is performed by combining material type identification and machine learning models, and the thickness calculation is corrected in real time to achieve dynamic model adaptation.
It enables adaptive detection of various substrates and copper foil materials, improves detection stability and adaptability, reduces environmentally dependent errors, has high adaptability and high-precision thickness prediction, reduces reliance on manual calibration, and improves detection efficiency and the reliability of quality management.
Smart Images
Figure CN121185188B_ABST
Abstract
Citation Information
Patent Citations
Real-time digitally enhanced imaging for the prediction, application, and inspection of coatings
CN106461373A
Typical target material attribute extraction method and device based on spectral video data
CN109872295A