Real-Time Fire Detection in Scenic Spot Using Convolutional Neural Network

• DOI: 10.20965/jaciii.2025.p0432
• Publication Date: 2025-03-19T15:02:19Z
• Authors: He Yan; Shaheem Sayed Mer...; Miao Zhang
• Keywords: Fire Detection

The current fire-detection methods rely primarily on smoke and temperature detection, which are generally performed in the late stage of fire thus cannot provide a timely reminder early fire. continuous development artificial intelligence has enabled machine-vision detection. This study proposes convolutional neural network target-detection algorithm, i.e., You Only Look Once version 4 (YOLOv4), to detect small targets. It offers outstanding characteristics enables scenic-spot monitoring via video extraction real-time detection using significant amount data. diverse scenes can accurate fire, providing favorable warning alarm function.