实验室火灾风险智能视觉预警仿真研究  

Simulation Research on Intelligent Visual Early Warning of Laboratory Fire Risk

作　　者：黄润樑[1] 王维正[2] HUANG Run-liang;WANG Wei-zheng(Dongguan University of Technology,Dongguan Guangdong 523000,China;College of Humanities&Information Changchun University of Technology,Changchun Hunan 130122,China)

机构地区：[1]东莞理工学院,广东东莞523000 [2]长春工业大学人文信息学院,吉林长春130122

出　　处：《计算机仿真》2023年第11期139-142,155,共5页Computer Simulation

摘　　要：当前的火灾风险视觉预警方法忽略了对火焰图像的增强处理,导致上述方法下火焰特征识别精度偏低。为此研究新的实验室火灾风险智能视觉预警方法。通过监控摄像装置实时采集图像,采用离散小波变换将采集图像分解成低通子带和高通子带图像,阈值去噪处理小尺度高通子带图像降低图像的噪声,增强图像对比度,获取增强后图像的火焰动态、静态特征。利用协方时空差融合火焰特征,经支持向量机分类器识别火焰图像后,通过实验室监控设备将识别的到火焰结果传输至微信APP完成视觉可视化预警。仿真结果表明,所提方法可有效降低图像噪声,提高火焰图像对比度,且静、动态特征融合后的图像特征识别率高达90%。所提方法可通过微信APP将预警信息实时传输给实验室负责人,实现实验室火灾风险的智能视觉预警。Currently,some methods ignore the enhancement of flame image,resulting in the low accuracy of feature recognition.Therefore,a new intelligent visual early warning method for laboratory fire risk is studied.The image could be collected in real-time by the monitoring camera.In the new method,the discrete wavelet transform was adopted to divide the collected images into low-pass subband and high-pass subband images.Then,the threshold denoising method was used to reduce the noise of the image and enhance the image contrast,thus obtaining the dynamic and static characteristics of the flame of the enhanced image.Moreover,flame characteristics were fused by using spatiotemporal covariance.After the flame image was recognized by the support vector machine classifier,the recognized results were transmitted to WeChat through laboratory monitoring equipment,thus completing visual early warning.Simulation results show that the proposed method can effectively reduce image noise and improve the contrast of flame images.Meanwhile,the feature recognition rate after static and dynamic feature fusion of images is as high as 90%.Through WeChat,the early warning information can be transmitted to the person in charge of the laboratory in realtime,thus realizing the intelligent visual early warning of laboratory fire risk.

关 键 词：实验室 智能视觉 火灾预警 对比度 分类器 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]