A micro-led defect detection method and system based on lightweight multi-scale convolution
CN117788413BActive Publication Date: 2026-10-16GUANGDONG UNIV OF TECH
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202311815955.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-27
- Publication Date
- 2026-10-16
- Estimated Expiration
- 2043-12-27
AI Technical Summary
Technical Problem
而基于卷积神经网络的目标检测方法仍存在一定的局限性，无法完全满足Micro-LED芯片的检测需求：
Benefits of technology
[0052] Firstly, the C2f module of YOLOv8 is replaced with a lightweight multi-scale convolution module, which enables the model to obtain rich gradient flow information while reducing the model size and computational cost.
✦ Generated by Eureka AI based on patent content.
Abstract
The application relates to the field of computer vision image processing, in particular to a Micro-LED defect detection method and system based on light-weight multi-scale convolution, and the technical scheme points are as follows: S1, a chip image acquisition system is constructed, which is used for shooting the chip image and generating a defect data set; S2, a neural network LBG-YOLO is constructed, which is used for defect positioning and classification of the chip. The light-weight convolution and the attention mechanism are used, the positioning precision of the network for the defect area is improved, the model size is reduced, and the calculation cost is reduced.
Need to check novelty before this filing date? Find Prior Art