视频监控图像斑点特征智能提取与仿真  

Intelligent Extraction and Simulation of Spot Features of Video Surveillance Images

作　　者：谭淅予[1] 叶丽娜[2] TAN Xi-yu;YE Li-na(Department of Criminal Science and Technology,Hunan Police Academy,Hunan Changsha 410000,China;School of Computer Science,Jilin Normal University,Jilin Siping 136000,China)

机构地区：[1]湖南警察学院刑事科学技术系,湖南长沙410000 [2]吉林师范大学计算机学院,吉林四平136000

出　　处：《计算机仿真》2023年第10期237-241,共5页Computer Simulation

基　　金：2021年吉林省预算内基本建设资金项目(2021C038-7)。

摘　　要：视频监控具有便捷、低成本的应用优势,但是摄像机获取的图像质量受环境影响较大,如当室外环境复杂或夜晚可视程度偏低时,很难及时获取监控异常情况。为此提出视频监控智能视觉图像斑点特征提取方法。利用智能视频监控系统采集视觉图像,通过图像灰度处理和图像二值化处理预处理采集到的视觉图像,将预处理后的视觉图像输入至视觉问答模型中,实现视频监控智能视觉图像斑点特征的提取。实验结果验证了所提方法提取的斑点特征与原图像具有一致性,可视化效果更好,且半点特征分布具有较高合理性,上述图像斑点特征提取方法应用下监控视频的图像分辨率更高,更利于监控视频的后续处理。At present,video surveillance is convenient and low-cost,but its image quality is greatly affected by the environment.If the outdoor environment is complex or the visibility is low at night,it is difficult to find the monitoring abnormalities in time.Therefore,an intelligent method of extracting spot features in the visual image of video surveillance was presented.First of all,an intelligent video monitoring system was used to collect visual images,and then these images were preprocessed by image grayscale and image binarization.After that,the preprocessed visual images were input into the visual question-answering model to intelligently extract the spot features of visual images in video surveillance.Experimental results prove that the spot features extracted by the proposed method are consistent with the original image.In addition,the visualization effect is improved,and the distribution of spot features becomes more reasonable.After the method is applied,the image resolution of the video is improved,which is more conducive to the subsequent video processing.

关 键 词：智能视频监控 视觉图像 视觉问答模型 斑点特征 图像灰度 图像二值化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]