摘要
隧道的半封闭空间结构对隧道的事故处置、安全运营管理带来巨大挑战。受图像处理算法和硬件发展的限制,传统的隧道监控视频车辆检测方法难以达到理想的检测效果,直接对隧道监控人员对事件的处置和决策带来极大影响。研究了基于毫米波雷达与机器视觉融合的隧道车辆目标检测技术,融合毫米波雷达和视觉传感的前端传感器为车辆目标检测提供了更为多源、准确的采集数据,并将获取的事件信息在融合处理后提供给隧道监控人员。该技术在保障隧道交通安全运营方面发挥重要作用。
Due to the semi-enclosed space structure of the tunnel, it is a great challenge to the traffic accident treatment and safety operation and management of the tunnel. With the restriction of image processing algorithm and hardware development, the conventional vehicle detection method by tunnel surveillance videos cannot obtain the ideal results, and has significant influence on the event treatment and decision-making of tunnel monitoring personnel. The tunnel vehicle target detection technology based on millimeter-wave radar and machine vision fusion was studied. The front-end sensor which combined millimeter-wave radar and vision sensing can provide more multi-source and accurate data collection, and the obtained event information was provided to tunnel monitoring personnel after fusion processing. This technology plays an important role in ensuring the safe operation of tunnel traffic.
作者
孙永安
SUN Yong-an(Yuncheng North Branch of Shanxi Transportation Holdings Group Co.,Ltd.,Yuncheng,Shanxi 044000,China)
出处
《山西交通科技》
2021年第4期66-69,共4页
Shanxi Science & Technology of Transportation
关键词
毫米波雷达
机器视觉
数据融合
车辆目标检测
millimeter-wave radar
machine vision
data fusion
vehicle target detection
