.png){kind=icon}
.png){kind=icon}
Return
Development and optimization of image fire detection on deep learning algorithms
.png){kind=icon}
PRE
AI
DOI：10.1007/s10973-022-11657-1.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The primary function of fire detection is to detect fires and raise the alarm early. A detection algorithm is a key element of image fire detection (IFD) technology because it directly determines the IFD's performance. In this study, an IFD algorithm based on the YOLOv3 network was developed to detect smoke and flame simultaneously. Subsequently, six improvements were applied to promote the algorithm's ability to detect fire early. The results demonstrated that the modified YOLOv3 network achieved an average accuracy of 95%, which is 14.1% higher than that of the same model without modifications. The detection speed reached 22 Frames Per Second (FPS), which satisfies the requirements of real-time detection.
Keywords:
Image fire detection
YOLOv3 network
Detection ability
Average accuracy
Detection speed
Journal
IF:
3.1
Papers:
1.8W
Citations:
3.2W