No icon

Smart Monitoring Cameras Driven Intelligent Processing to Big Surveillance Video Data

Smart Monitoring Cameras Driven Intelligent Processing to Big Surveillance Video Data


Video surveillance system has become a critical part in the security and protection system of modem cities, since smart monitoring cameras equipped with intelligent video analytics techniques can monitor and pre-alarm abnormal behaviors or events. However, with the expansion of the surveillance network, massive surveillance video data poses huge challenges to the analytics, storage and retrieval in the Big Data era. This paper presents a novel intelligent processing and utilization solution to big surveillance video data based on the event detection and alarming messages from front-end smart cameras. The method includes three parts: the intelligent pre-alarming for abnormal events, smart storage for surveillance video and rapid retrieval for evidence videos, which fully exploresthe temporal-spatial association analysis with respect to the abnormal events in different monitoring sites. Experimental results reveal that our proposed approach can reliably pre-alarm security risk events, substantially reduce storage space of recorded video and significantly speed up the evidence video retrieval associated with specific suspects.

Existing System:

The existing intelligent surveillance systems can on-ly detect and alarm single abnormal event yet without bridgingthe spatial and temporal association among mul-tiple unusualevents. However, it is quite not convincible to judge suspicious behavior by a single monitoring. As the case of wandering in the front of a bank, the occasion-al wander outside the bank may be a usualbehavior for awaiting others. It only makes sense to treat the wander-ing as suspicion when it happens repeatedly or takes a long time.



Proposed System:

The recent emergingsmartmonitoring camerasare able to automatically identify abnormal behaviors through the built-in intelligent algorithms, greatly boost-ingthe performance of the surveillancesystem.However, theabove mentioned three major challenges have not been fundamentally resolved. The essentialreason is that the existing systemonly individually acceptsalarm in-formation fromeach front-end camera and makesa lim-ited range of alarming, without performing collaborativeanalysisamonggeospatiallyinterrelated camera network. Besides, the detection results on unusualbehaviors are not fully exploited in terms of deep utilization, paying little attention to storage and retrievalonmassivevideobut for event alarming.

This paper advocates anintelligent processing approach to big surveillance video datadriven by smart front-endcameras. In our approach, we do not nativelyand passivelyreceive and process the alarminginformationfromsmart front-endcameras, but make full use of spatial and temporal attributes of multi-sitemoni-toring camerasto perform collaborative association analysis. This approachwill disclosethe intrinsic relation-ships and reveal hidden patterns among a numberof seemingly separate abnormal behaviors.In methodology, it is applied to three major procedures of intelligent video surveillance system: danger alarming before an event, high-efficiencystorage duringan event, and rapid evi-dence retrieval after an event. This way, we canimprove thealarmingaccuracyofthe abnormal behaviorswithinherent association, the efficiency of the video preserva-tion associated with the abnormal behaviors, and the dis-covery efficiency of the caseclues under the abnormal behavior constraint.

Comment As:

Comment (0)