Research on Gas Pipeline Fire Detection Algorithm Based on Improved YOLOv8
The improved gas pipeline fire detection algorithm based on YOLOv8n plays a crucial role in the safety monitoring and fire warning system of gas pipelines, enhancing the detection rate and response speed of fire, effectively reducing human and financial losses caused by fire. This study established...
Saved in:
Published in | 2024 IEEE 7th International Conference on Electronic Information and Communication Technology (ICEICT) pp. 1298 - 1301 |
---|---|
Main Authors | , , |
Format | Conference Proceeding |
Language | English |
Published |
IEEE
31.07.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The improved gas pipeline fire detection algorithm based on YOLOv8n plays a crucial role in the safety monitoring and fire warning system of gas pipelines, enhancing the detection rate and response speed of fire, effectively reducing human and financial losses caused by fire. This study established a dataset of fire images, incorporated the BiFPN network into YOLOv8n, and introduced the Focal DIoU Loss for bounding box regression calculation. Compared to the unimproved YOLOv8n, this algorithm has achieved an increase in Precision by 0.16%, Recall rate by 3.36%, and mAP@0.5 by 1.36%. |
---|---|
ISSN: | 2836-7782 |
DOI: | 10.1109/ICEICT61637.2024.10671153 |