Дякуємо, ми отримали ваше повідомлення і звʼяжемось в найближчий час! :)
How it works:
● Scylla AI video analytics detects people or vehicles within a defined area during a specified time window.
● Scylla easily integrates with existing security infrastructure, connects directly to cameras, and monitors the premises in real time, 24/7.
● Additionally, it can receive frames triggered by built-in motion detection, filter those that contain a person or vehicle, and send relevant alerts upon detection.
● Scylla can track individuals while they remain in the camera’s field of view, count vehicles or people on your premises in real time, and notify you about movement or running events within designated zones.
What does Scylla AI Object Detection and Tracking recognize?
The system is trained to detect and identify a wide range of weapons, knives, robbery masks, and unattended objects.
Robbery masks
Foreign objects, debris
Unattended items
Litter detection
AI-Powered Technologies for Public Safety and Risk Reduction
Robbery Mask Detection
In situations where individuals attempt to conceal their identity during a robbery using masks or balaclavas, traditional surveillance methods may fail to provide timely and accurate threat assessment. Automated detection technology leverages advanced algorithms to analyze video streams in real time, identifying the presence of robbery masks or balaclavas.
Foreign Object Debris
Foreign object debris (FOD) at airports can cause substantial damage, costing airlines and airports millions of dollars annually. These objects may be found near terminal gates, cargo ramps, runways, and other areas throughout the airport. It’s estimated that FOD causes up to $4 billion in losses to the aerospace industry each year, not to mention potential aircraft damage and injuries.
Scylla’s AI-powered Object Detection can effectively analyze footage from high-quality cameras, detect a wide range of unattended or misplaced items from a distance, and send alerts about potential FOD to all designated endpoints
Abandoned Object Detection
Abandoned object detection (AOD) is a critical concern as transportation hubs, critical infrastructure facilities, and law enforcement agencies aim to leverage their video surveillance networks to identify unattended bags in public spaces—both in real time and for investigative purposes. Abandoned items can pose serious risks to public safety and require immediate attention from police and security professionals.
Scylla’s proprietary object detection algorithms can identify abandoned items and notify security personnel as soon as they are detected. This enables timely response and helps save valuable time.
Litter Detection
Litter in public areas, on roads, and at transportation hubs poses not only an environmental hazard but can also lead to financial losses.
Most such items are difficult to identify using conventional AI. However, Scylla leverages robust object detection algorithms capable of spotting small, unattended items from a distance—even against dynamic or moving backgrounds.
What sets the Scylla Object Detection and Tracking System apart from other solutions?
● AI methodologies operate autonomously 24/7 and continuously self-improve
● Seamless integration with most cameras and video surveillance systems
● Performs effectively on cameras with dynamic backgrounds, such as drones and bodycams
● Employs a built-in zooming and tracking algorithm, enabling detection of distant objects
● Requires 7x less hardware compared to similar solutions on the market
● Can be deployed both on-premises and in the cloud
● The object tracking module is built on a proprietary algorithm that is lightweight, accurate, and versatile
● The AI model is custom-designed for re-identification of specific object types with unmatched accuracy
● The algorithm is easily scalable for centralized deployments analyzing hundreds of video streams simultaneously
FAQ
Can the Scylla Intrusion Detection System be integrated into an existing video surveillance network?
Yes. The Scylla Intrusion Detection and Perimeter Protection System is one of the most flexible solutions available, as it can be easily integrated with most video surveillance and recording systems.
Depending on the infrastructure already in place at the site, integration can be either one-way or two-way:
In a one-way integration, Scylla IDS consumes video streams directly from existing cameras or from NVR/DVR devices they are connected to. The analysis results are then displayed in the Scylla dashboard.
In a two-way integration (e.g., with major VMS providers like Mobotix, Genetec, NX Witness, etc.), the analysis results can be viewed directly within the corresponding VMS dashboard.
What is the detection response time?
Typically, it is less than one second. In cloud deployments, the response time may slightly increase depending on the client’s upload speed.
What objects can the Scylla Intrusion Detection and Perimeter Protection System detect?
Currently, the Scylla Intrusion Detection System is capable of detecting people and several types of vehicles, including passenger cars, trucks, buses, vans, and motorcycles.
Does Scylla Intrusion Detection and Perimeter Protection comply with GDPR and CCPA requirements?
Absolutely. Scylla does not store any data that could be considered personal. We do not retain any video recordings or images. The only data that may be stored are alert notifications, and their retention period can be adjusted according to the policies set by the client.
Is there a cloud version of the Scylla Intrusion Detection and Perimeter Protection System?
Yes, there is. In fact, nearly all Scylla products can be deployed in the cloud. In a cloud-based scenario, video or image frames from surveillance cameras are transmitted to the cloud using one of the following methods:
Via Scylla Connector software – proprietary software that connects to the camera, receives the video stream, encodes it, and sends it packet-by-packet to the cloud, where decoding and further analysis occur. This architecture is ideal for environments with local network constraints (e.g., domain, port restrictions).
Using built-in camera algorithms – such as motion detection, which sends one or more frames or short video snippets to the cloud via HTTP/FTP or similar protocols. This scenario is preferred when the client wants to minimize bandwidth usage. It is commonly used in Scylla’s False Alarm Filtering system.
Direct camera-to-cloud connection – the least popular method due to associated technical complexities and security concerns.
What are true and false alarms?
An alarm is classified as true when the AI's prediction matches reality—for example, when an object of interest is correctly identified or a specific action is accurately detected. A false alarm occurs when an alert is triggered mistakenly.
Due to the probabilistic nature of AI, false positives are largely unavoidable. However, thanks to the advanced artificial intelligence and machine learning behind the Scylla Perimeter Intrusion Detection System—as well as built-in proprietary algorithms like “Eagle Eye”—the system meets any industrial-grade standard.
Moreover, Scylla AI’s video analytics modules are continuously improved through retraining on past errors, which helps reduce the number of false alarms even further over time.
Does the system prevent intrusions?
The Scylla Intrusion Detection and Perimeter Protection System alerts the appropriate security team or personnel about an intrusion event. The alert includes valuable visual and metadata about the origin of the intrusion.
Additionally, the intruder detection system can be integrated with Access Control Systems (ACS) to help prevent further breaches by locking down infrastructure, triggering alarms, disabling entry protocols, and more.
Is there a "blacklist" feature in case I want to detect only those not on a whitelist?
Yes, if the Scylla Facial Recognition Module is deployed alongside the Perimeter Intrusion Detection System, it can be used to enroll individuals detected on the premises into whitelists and watchlists.
However, since the facial recognition module is a separate solution and not mandatory for the Scylla Perimeter Intrusion Detection System, there are several important considerations:
Additional hardware is required to run the facial recognition module. It typically needs around 2 GB of extra GPU memory and additional processing power.
The accuracy of watchlist recognition heavily depends on external factors such as facial visibility, face size, angle, lighting, and potential obstructions. Unlike intrusion detection, which can be triggered by any visible part of a person, facial recognition is more demanding.
To improve identification accuracy, the perimeter intrusion detection should be deployed together with person tracking, which increases the likelihood of capturing biometric data when the face becomes visible. However, person tracking is computationally intensive and also requires additional hardware.
As mentioned, the facial recognition module is a standalone product and must be purchased separately.
Fill out the form to receive a personalized consultation for your IT infrastructure: