We have heard that it is quite common to get hundreds or thousands of false alarm events from your security system. This is a big distraction for customers.
We want our customers to be able to focus on their business. There is no need to react to false alarms. Reconeyez uses deep neural network-based technology
to classify received photos into different categories (person, car, truck, bus, motorcycle, bicycle, dog, horse),
confidence (How sure the model is that the detected object belongs to the assigned category?),
estimated distance of detected object from device,
detected object direction estimate,
detected object speed estimate.
Our patent pending technology tries to assign category to all the received photos. If no category is found on the photo, then the photo is automatically considered as dismissed (false positive) event.
The above image illustrates a real-life situation from our customer's site. The blue line shows an amount of alarm events per day received from that site. Grey (not alarms) and red (alarms) lines are automatically assigned by Reconeyez' patent pending technology.
Even if the Reconeyez Detector detects and sends 4-24 events per day the customer is getting only a couple of possible alarm incidents per day (except on 14.June when there was lots of activity on site).
The example screenshot above shows an alarm incident of two persons entering the construction site. Reconeyez automatically marked the event as an alarm. Reconeyez system also adds tags on the detected objects (see image above). You can toggle on/off to see the detected objects on an alarm image.
You can change your security business with Reconeyez because it allows you to only focus on the important information.