基于阈值化的开放式无人值守实验室安防监测  

Open Unmanned Laboratory Security Monitoring Based on Thresholding

作　　者：赵飞 贾和坤[2] ZHAO Fei;JIA He-kun(School of Automotive Engineering,Changshu Institute of Technology,Changshu Jiangsu 215500,China;School of Automotive and Traffic Engineering,Jiangsu University,Zhenjiang Jiangsu 212000,China)

机构地区：[1]常熟理工学院汽车工程学院,江苏常熟215500 [2]江苏大学汽车与交通工程学院,江苏镇江212000

出　　处：《计算机仿真》2024年第11期296-300,共5页Computer Simulation

摘　　要：为了确保开放式无人值守实验室的安全运行,提出一种开放式无人值守实验室远程安防监测方法。将实验室监测窗口划分为四个子窗口,将计算获取的加权平均值和预先设定的阈值展开比较,准确区分噪声点以及信号点,引入边缘保持滤波方法重构被污染的噪声像素灰度值。在完成滤波处理后,引入改进背景差分法提取开放式无人值守实验室视频图像中的背景区域,将视频流中相邻两者图像的像素相减,提取实验室图像前景,对相减后的图像展开阈值化处理,标记出开放式无人值守实验室的运动区域,实现开放式无人值守实验室远程安防监测。通过仿真分析证明,采用所提方法可以得到更加精准的监测结果,同时还可以有效减少监测延时。In order to ensure the safe operation of open unmanned laboratories,a remote security monitoring meth-od for open unmanned laboratories was proposed.The method first divided the monitoring window into four sub-win-dows,and compared the weighted mean value with the preset threshold,and thus to accurately distinguish noise points from signal points.Then,an edge protection filter algorithm was introduced to reconstruct the gray values of polluted noise pixels.After the filtering process,an improved background difference method was used to extract the background area in the video of the open unattended laboratory.Moreover,the foreground of laboratory image was extracted by the subtraction between the pixels of adjacent images in video stream.After that,the image was thresholded.Meanwhile,the motion area of the open unmanned laboratory was marked.Finally,the remote security monitoring was realized.Through simulation analysis,it has been proven that the proposed method can obtain more accurate monitoring results while effectively reducing monitoring delays.

关 键 词：开放式 无人值守实验室 远程安防监测 改进背景差分法 噪声滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]