【摘要】 为了能够在不丢失有用信息的前提下对海量监控视频进行摘要,提出一种基于目标检测及跟踪的视频摘要技术。首先根据混合高斯模型对视频进行背景建模,获得监控视频的背景;然后通过背景减法对运动目标进行检测,得到前景运动目标;其次对检测到的运动目标采用分层关联的思想进行目标跟踪,得到运动目标的完整信息;最后将这些运动目标及视频背景重新组装成摘要视频。实验结果表明,该方法能够有效地对监控视频进行浓缩,形成的摘要视频能够完整地保存原有视频信息,减少了存储空间,降低了成本,方便了相关人员及时获取有用信息,提高了工作效率。更多还原

【Abstract】 In order to carry out a summary of massive surveillance video under the premise of not losing useful information,a video summarization technique based on object detection and tracking was proposed.Moving targets is detected by background subtraction in video by using Gaussian mixture model,and target tracking for detected target is done by using the idea of hierarchical correlation to get complete information of moving targets.Finally,the moving objects and video background are reassembled into the abstract video.The experimental results show that the method proposed in this paper can be used to concentrate the monitoring video.The summary of the video can completely preserve the original video,which reduces the storage space and cost.It is also convenient for relevant personnel to obtain useful information in time and improve work efficiency.更多还原