基于视频图像的火焰特征分析  

Analysis of Flame Features Based on Video Images

作　　者：邵坤艳 刘丹丹 余迪 SHAO Kunyan;LIU Dandan;YU Di(Shangqiu Normal University,Henan 476000,China.)

机构地区：[1]商丘师范学院,河南476000

出　　处：《电子技术(上海)》2025年第1期86-88,共3页Electronic Technology

基　　金：河南省教育厅2025年度河南省高等学校重点科研项目(25A510022)。

摘　　要：阐述在基于视频图像的火灾检测方法中,火焰特征提取至关重要。为此,分析火焰的静态特征和动态特征,建立火焰的颜色模型,提取了疑似火焰区域。通过运动目标检测,排除了类似火焰颜色的物体干扰。通过提取火焰的闪烁、面积变化等动态特征,进一步判断待检测区域是否发生火灾。This paper describes the importance of flame feature extraction in fire detection methods based on video images.It analyzes the static and dynamic characteristics of flames,establishes a color model of flames,and extracts suspected flame areas.It eliminates interference from objects similar to flame colors through motion target detection.It further determines whether a fire has occurred in the area to be detected by extracting dynamic features such as flame flickering and area changes.

关 键 词：视频图像 计算机视觉识别 颜色特征 动态检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]