论文部分内容阅读
随着视频监控系统在公安、交通等敏感领域的大规模应用,电子眼帮助破案的报道时有所耳闻。一方面,这些事例从正面体现了视频监控系统的社会价值;另一方面,也引发了更多的关于视频检索的深层次需求。在实际应用的过程中,用户常常希望可以快速的从海量的数以万计的摄像头视频录像中,方便的找到一些有明显特征的人或物。传统的基于预设告警和时间的视频检索方式,对于这类深层次的需求往往无能为力,常有“隔靴搔痒”之憾,因而如何快速准确的进行智能的检索
With the video surveillance system in public security, transportation and other sensitive areas of large-scale application, the electronic eye help solve the reported case when heard. On the one hand, these examples show the social value of video surveillance system from the front; on the other hand, they also lead to more deep-seated demands on video retrieval. In the practical application process, users often want to be able to quickly find the obvious people or objects from tens of thousands of massive video cameras. The traditional method of video retrieval based on the preset alarm and time often can not help such deep-seated needs, and often has the problem of “scratching it”, so how to quickly and accurately carry out intelligent retrieval