Video Guard improves construction site safety
Artificial intelligence (AI) is now simplifying human activities in many areas of life. This applies to both private and professional contexts. The use of AI is also proving to be extremely beneficial in the construction industry, for example in securing and monitoring construction sites. The camera towers from Video Guard, for example, use self-learning video analytics to secure their surveillance areas in the best possible way. The AI-supported detection analysis helps with surveillance and scores points with its reliable and fast operation.
In the construction industry, the use of artificial intelligence is generally diverse: it provides support in the procurement of new construction projects, in project management or directly in the implementation of construction projects on site. On construction sites, it helps to improve occupational safety by reducing extreme physical strain and preventing accidents at work. It can also save construction time and costs. By using AI in video surveillance, the question of which changes in the surveillance area are okay and which pose a real danger can be answered more quickly and reliably. Dr. Benedict Doepfer, Vice President Sales at International Security GmbH, sums up the advantages: “Video Guard detects quickly and clearly. If necessary, intervention is immediate.”
Without AI: sensitive cameras cause many alarms
With conventional camera analytics without AI, even small pixel changes can trigger alarms. As a result, the actual alarms are always mixed in with numerous other alarms, which can cause excessive demands on the control center. To prevent this, the alarm would have to react much less sensitively. However, this could in turn lead to an alarm being delayed or possibly not triggered at all, even in the event of actual threats from intruders.
Self-learning video analytics for reliable surveillance
A second layer of cloud AI video analytics in addition to camera analytics enables Video Guard to provide truly reliable video surveillance in the first place. In this case, more computing power allows the use of better analysis models. As a result, alarms caused by conventional camera analytics can be significantly reduced. This is because the AI reliably learns to differentiate between normal movements, such as animals crossing the premises, and actual dangers posed by potential thieves or vandals. By reducing the number of alarms, a single employee can monitor significantly more cameras, which makes work in the control center much easier.
Ongoing detection analysis
The connected control center is staffed around the clock. If it is informed by the system of an unauthorized entry into the surveillance area, it can therefore evaluate the image material immediately. If necessary, the intruders can be addressed directly and loudly via the loudspeakers on the Video Guard towers. If this does not drive the uninvited visitors away, further measures such as alerting the police are taken. On average, it takes just two minutes between unauthorized entry to the premises and a corresponding response from the control centre. The rapid response is made possible above all by the AI video analytics, which simplify the decision as to whether the control center needs to take action or not. Thanks to the rapid intervention, the chances of finding the intruders on site and preventing damage caused by theft or vandalism are significantly higher.
