The sheer volume of video that is being captured by public safety agencies and governments at all levels is far beyond anyone’s ability to view and to gain meaningful information. Yes, it is useful for reference after a crime or major incident, but there is so much more that could be done to make beneficial use of what has been photographed.
Only the tiniest fraction of video surveillance footage is ever reviewed later. That is why automating the video analysis process and infusing it with artificial intelligence is the key not only to quick response but also information gathering to enhance security and public safety in government buildings and venues.
Smart video surveillance analytics allow organizations to detect anomalies in a video in real time, in order to trigger security alerts and enable rapid action. For instance, if someone leaves a parcel in city hall, a state government building, or a federal facility and walks away, video analytics can catch that instantly. Or if a door is left ajar, the camera would capture it, but it might escape human detection. An artificial intelligence-based analytics system can spot that and issue an immediate security alert.
How are video analytics valuable? Here are just a few examples.
- Intrusion detection: With advanced video surveillance analytics, you can immediately detect when people or vehicles enter restricted areas. The system will generate an alarm and a security alert if an object moves through a pre-defined area of the video in a given direction. This is highly valuable in areas where entries, doors, windows, fences, or walls need to be monitored – areas such as prisons or jails, government buildings, or municipal facilities such as water plants or power substations.
- Object detection: This detects changes in critical areas such as interlocks, elevators, laboratories, or museums, and automatically sends alarms and security alerts. Potentially dangerous objects such as left-behind luggage or weapons can be sensed immediately to enable removal by security personnel. Even tiny objects, such as lighters or pocket knives, can be sensed, and the system can also detect whether an object has been moved or removed from its original position.
- Facial detection: A video analytics system can be highly effective at detecting and logging faces in a video stream. For example, it can aid police in searching for criminal or terrorist suspects or in missing person investigations. It can enable trespass security alerts in order to deny entry into certain areas to specific individuals such as past offenders. It can also enable identity verification for fraud prevention or loss prevention alerting for known offenders in retail or other environments.
- License plate recognition: This automatically recognizes vehicle license plates and matches them with whitelisted or blacklisted plates. Its use by police agencies for finding at-large suspects or stolen cars is already well documented, but it also has value in monitoring areas such as international borders or toll collection sites.
- Traffic sensing: The video surveillance analytics solution can automatically count and classify vehicles on roads and intersections. That includes the ability to instantly detect whether a driver has entered a highway going the wrong direction, a particularly dangerous event, and hopefully allow police to intercept that driver before a collision occurs.
- Video optimizer: Sometimes due to weather or other factors, the quality of a video may be reduced. The video analytics system can overcome that lower visibility through intelligently filtering and stabilizing the image, to improve its quality for both real-time video analysis and later review.
- Camera health monitor: This can detect camera tampering as well as connection problems and poor picture quality, triggering an alarm and a security alert so that the situation can be assessed and repairs made as needed. This is important in areas where the availability and integrity of video surveillance cameras are a high priority.
Getting the 3D view
A fascinating aspect of video analytics is its ability to use Lidar to enhance the available information. Lidar is similar to radar, but uses laser light to visualize an object three-dimensionally. Beyond knowing that an individual was captured on the video feed, it can tell you precisely how tall that person is and give you a more comprehensive view of that person’s approximate body structure, clothing, and more.
Using that example of a person leaving a parcel behind and walking away from it, with Lidar you don’t just get a fleeting view of that person, you get a full image of them, which makes it far easier to leverage video analytics and other cameras throughout the venue to locate that person or determine if they have left the building. That allows instantaneous determinations of whether something might be a threat or reveal a simple honest mistake.
By using video analytics and Lidar in a building – perhaps a courthouse or larger city, state, or federal government facility, it is possible to know at any given time how many individuals are in that building.
In the event of an incident such as a shooting or a fire, it is crucial that everyone be evacuated immediately. Video analytics and Lidar allow you to better determine which individuals have been evacuated and who might still be in the building. It isn’t enough to know that there were 150 people in that building at the time of the event, and only 148 have been evacuated. By knowing which individuals are still in there, security personnel know who to look for.
Making it easy to use
A video surveillance analytics solution such as Sprint’s Curiosity Smart Video Analytics is designed to be both easy to use and cost-effective for public safety and government agencies.
Rather than require a costly up-front equipment investment, Curiosity Smart Video Analytics is offered as a service. Government agencies don’t need to worry about issues such as how to store all the video, how to find the necessary skilled staff, or how to afford the ongoing operations and maintenance costs once the potentially expensive equipment is in place. It is available via a monthly operational charge that is easy to budget for.