Дякуємо, ми отримали ваше повідомлення і звʼяжемось в найближчий час! :)
How It Works:
● Scylla’s proprietary object detection algorithm is trained to identify a set of predefined objects in one or multiple frames. The initial detection is fast and serves to flag potential interest areas for more precise evaluation by the Charon classifier.
● Following this, an alert is triggered and sent to the end user via Scylla’s web channel, mobile app, or integrated VMS channels, including the time, location, and a screenshot of the detection.
● Scylla object detection is independent of background scenes or motion and can accurately analyze video content from both stationary and moving cameras.
● Scylla Object Tracking monitors the movement of objects across a sequence of video frames.
● Tracking the same object across multiple consecutive frames allows for repeated evaluation, further enhancing the accuracy of Scylla AI physical security decisions in real-time video analytics.
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 abandoned objects:
Robbery masks
Foreign objects, debris
Abandoned items
Litter detection
AI-Based Technologies for Public Safety and Risk Reduction
Robbery Mask Detection
In situations where individuals attempt to conceal their identity with masks or balaclavas during a robbery, traditional surveillance methods may fail to provide timely and accurate threat assessment.
The automatic detection technology uses advanced algorithms to analyze video streams in real time, identifying the presence of robbery masks or balaclavas.
Уламки сторонніх предметів
Уламки сторонніх предметів (Foreign object debris. FOD) в аеропортах можуть спричинити значні збитки, які щороку коштують авіакомпаніям та аеропортам мільйони доларів. Сторонні предмети можна знайти біля воріт терміналів, вантажних перонів, злітно-посадкових смуг та інших місць в аеропорту. За оцінками, збитки від них обходяться аерокосмічній галузі в 4 мільярди доларів на рік, не кажучи вже про можливі пошкодження літаків і травми.
ШІ для виявлення об'єктів Scylla може ефективно аналізувати відео з високоякісних камер, виявляти різноманітні об'єкти, що залишаються без нагляду, на відстані та надсилати сповіщення про можливі уламки сторонніх об'єктів на всі призначені кінцеві точки.
Foreign Object Debris
Foreign object debris (FOD) at airports can cause significant damage, costing airlines and airports millions of dollars annually. Such debris can be found near terminal gates, cargo aprons, runways, and other airport areas. It is estimated that FOD costs the aerospace industry $4 billion per year, not to mention potential aircraft damage and injuries.
Scylla’s AI object detection can effectively analyze video from high-quality cameras, detect various unattended objects at a distance, and send alerts about potential foreign object debris to all designated endpoints.
Litter Detection
Litter in public places, on roads, and in transportation hubs not only poses an environmental hazard but can also cause material damage.Most such items are difficult to identify using conventional AI. However, Scylla employs robust object detection algorithms that enable it to spot small unattended objects at a distance, even against moving backgrounds.
What Sets the Scylla Object Detection and Tracking System Apart from Other Solutions?
● AI methodologies operate autonomously 24/7 and self-improve
● Easily integrates with most cameras and video surveillance systems
● Performs effectively on cameras with moving backgrounds, such as drones and body-worn cameras
● Uses a built-in proximity and tracking algorithm that enables detection of distant objects
● Requires 7 times less hardware resources compared to similar solutions on the market
● Can be deployed both on-premises and in the cloud
● The object tracking module is based 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 can be easily applied in centralized solutions analyzing hundreds of video streams simultaneously
Fill out the form to receive a personalized consultation for your IT infrastructure:
FAQ
Can the Scylla Object Detection System Work Effectively Outdoors?
Yes, the systems are trained on a wide variety of landscapes and backgrounds, under different lighting conditions and from various angles. Essentially, the solution is highly resilient to background variations — as long as the object of interest is visible, the system will reliably detect it.
What is the Maximum Detection Distance?
The answer depends on several factors, primarily the camera’s specifications, particularly its resolution. Resolution plays a significant role initially; however, Scylla’s built-in scaling and tracking algorithm allows objects to be analyzed at the camera’s original resolution. Unlike many AI-based security solutions, Scylla is less dependent on image quality, which often degrades during processing on neural network platforms.
There are several factors associated with “image quality” such as stream bandwidth, encoding, and so on. Visibility conditions like lighting, object orientation (see question 6 about object tilt angle), and the object’s pixel size also matter. The latter depends linearly on the distance to the camera and can be used to estimate the maximum detection range.
For example, the reliable minimum object size for weapon detection is approximately 15–17 pixels, which translates to a maximum distance of about 10–12 meters for most HD cameras.
What is the Detection Response Time?
Detection typically occurs within the first 400 milliseconds (in some cases up to 2 seconds). When evaluating response time, consider that most IP cameras in use today introduce some sub-second video stream delay. Additionally, when Scylla is deployed in the cloud, the latency of the video stream reaching the cloud and the response returning to the dashboard should also be taken into account.
How Does the Scylla Object Detection System Assist Law Enforcement?
The Scylla Object Detection System is designed to support security units by enhancing their daily operations, expanding their capabilities, and eliminating potential shortcomings related to human factors. Furthermore, in the event of a possible threat, the alerts issued by Scylla are enriched with information critical for rapid and comprehensive threat analysis on-site, enabling effective planning of specialized countermeasures.
Can the system detect a weapon if someone is carrying it in a bag?
The system relies on computer vision algorithms, and threat detection is based on visual content analysis. This means that to detect weapons inside bags, Scylla must be integrated with X-ray or millimeter-wave scanners. When operating on video surveillance cameras that work only in the visual spectrum, Scylla Object Detection can detect only exposed weapons.
Does the object need to be at a specific angle for the system to detect it?
No, the system is trained to recognize objects from all possible angles. However, in some specific cases, the angle of the object may matter because the features that Scylla uses to classify the object are clearer from certain angles than others. For example, if a pistol or rifle is held at an angle to the camera, more distinct features may be visible compared to when it is pointed directly at the camera.
What does the system do after detection?
An alert containing all relevant information is generated and sent to the security personnel responsible. There are several configurable notification channels, including the Scylla dashboard, Scylla mobile app, relay boards at access points, and VMS alert APIs, among others.
Can Scylla Object Detection operate in the cloud?
Yes, all Scylla solutions can be deployed both in the cloud and on-premises. Moreover, Scylla’s AI software solutions are cloud-provider agnostic, as long as the cloud instance runs Linux and is equipped with an Nvidia GPU.
Can the object detection system work with infrared/thermal cameras?
Yes, it can. The maximum detection distance of the Scylla Object Detection Solution depends on the camera’s characteristics (contrast ratio, pixel crosstalk, etc.). Overall, the solution meets standard industry DRI requirements, with an identification threshold (the distance at which the object class can be determined) of approximately 20 pixels for firearms.
What are True Positives and False Alarms?
An alarm is classified as a true positive when the AI prediction corresponds to reality (i.e., the object of interest is correctly identified, the targeted action is detected, etc.). A false alarm occurs when an alert is triggered mistakenly. Unfortunately, due to the probabilistic nature of AI, false alarms are often inevitable.
However, thanks to the advanced artificial intelligence and machine learning underlying the Scylla Object Detection system, it can meet any level of industrial-grade standards. Moreover, we continuously improve Scylla AI video analytics modules by retraining them on errors, which gradually reduces the number of false alarms over time.
Recommended Camera Specifications
A simple empirical rule applies to most questions about camera limitations and requirements: if a person can see and identify the object of interest, then the Scylla AI video analytics system can also do so — and in some cases may even outperform humans thanks to built-in scaling algorithms and repeated verification.
Regarding minimum camera specifications, these depend on the specific use case and the object you are trying to detect. Of course, the camera should have a digital output or at least be connected to a video recorder that provides one. The Scylla Object Detection system can accept almost all stream types, such as RTSP/RTMP, HTTP, etc.
Typically, the minimum required resolution starts from HD (1280x720) at 5 FPS. Parameters determining frame/image quality vary from camera to camera, but we recommend paying attention to characteristics such as bandwidth, encoding, and sharpness and improving them as needed.
Does the Scylla Object Detection System comply with GDPR and CCPA?
Absolutely. Scylla does not store any information unless specifically requested by the user.
Can the solution operate in harsh conditions?
The Scylla Object Detection system is designed to perform in challenging environments where cameras with built-in algorithms may not function properly. The AI engine compensates for limitations caused by difficult conditions, including poor lighting, somewhat distorted frames, environmental factors, and weather effects.