摘要
图像自动识别技术是计算机图像专业的新型研究领域,同时也是饱受人们关注的学术课题。传统监控视频主要是依据肉眼可视分辨,需要消耗大量人力和物力。基于此,提出图像自动识别技术在行人异常行为检测中的应用。通过图像自动识别技术在行人异常行为检测的一般流程,在划定监测区域的前提下,进行信息扫描识别基础上的特征提取,最终将检测信息和追踪信息进行整合。
Automatic image recognition technology is a new research field of computer image specialty, and it is also an academic subject which attracts people＇s attention. Traditional surveillance video is mainly based on the visual discrimination of the naked eye, which requires a lot of manpower and material resources. Based on this, the application of image automatic recognition technology in the detection of pedestrian abnormal behavior is proposed. Through the automatic process of image recognition, the general process of pedestrian abnormal behavior detection, the premise of delineating the monitoring area, the feature extraction based on information scanning and identification, and finally the integration of detection information and tracking information.
作者
陈纪铭
陈利平
Chen Jiming;Chen Liping(Hunan Institute of Technology,Hengyang Hunan 421002,China)
出处
《信息与电脑》
2018年第22期144-145,共2页
Information & Computer
关键词
图像自动识别技术
行人异常行为
检测
跟踪
automatic image recognition technology
pedestrian behavior
detection
trac
