Mura defect detection of LCD screen based on improved YOLOv8n

Table 1. Description of the experimental environment
View table
View in Article
Table 1. Description of the experimental environment
软件硬件
Pytorch torchvision Python CUDA GPU CPU 内存
1.11.0 0.12.0 3.9.19 11.7 NVIDIA RTX 3060Ti 8G AMD 5700X@3.40 GHz 32G

Table 2. Ablation experiments
View table
View in Article
Table 2. Ablation experiments
方案 C2fConX CSDH NWD GFLOPs P/% R/% mAP₅₀/%
A 8.1 89.5 86.1 91.2
B √ 6.9 91.9 86.6 93.2
C √ √ 6.1 92.6 85.4 93.4
D √ √ √ 6.1 92.9 88.8 94.8

Table 3. Performance comparison of different algorithms
View table
View in Article
Table 3. Performance comparison of different algorithms
模型 GFLOPs FPS P/% R/% mAP₅₀/%
YOLOv5n 4.2 109.9 90.8 85.7 88.8
YOLOv5s 15.8 99.0 92.4 87.0 92.8
YOLOv8n 8.1 97.1 89.5 86.1 91.2
YOLOv8s 28.4 85.4 93.6 87.7 93.7
YOLOv10n 8.2 76.9 90.5 84.2 90.4
YOLO-D3MNet 6.1 92.5 92.9 88.8 94.8

Tools

Get Citation

Copy Citation Text

Shunlong CHEN, Yinghua LIAO, Feng LIN, Chengye SHU. Mura defect detection of LCD screen based on improved YOLOv8n[J]. Chinese Journal of Liquid Crystals and Displays, 2025, 40(3): 439

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category:

Received: Sep. 24, 2024

Accepted: --

Published Online: Apr. 27, 2025

The Author Email: Yinghua LIAO (liaoyinghua118@163.com)

DOI:10.37188/CJLCD.2024-0295

Topics

Table 1. Description of the experimental environment

Table 1. Description of the experimental environment

Table 2. Ablation experiments

Table 2. Ablation experiments

Table 3. Performance comparison of different algorithms

Table 3. Performance comparison of different algorithms