Types of Analytics

Video Analytics within the system is categorized into three levels, reflecting varying degrees of processing complexity. The classification is based on the algorithms applied and the tasks they address. This structured approach clarifies the system capabilities and enables users to quickly select the most appropriate tools. The levels range from basic event detection to advanced identification powered by reference models.

In this article:

• General description of each level

• Types of Analytics included in each level

• Events that are detected by the system for each Analytics type 

Analytics. Level 1

Analytics solutions at Level 1 offer basic video and audio detection tools that automatically track important events and conditions. Level 1 includes:

• Detection of loud sounds

• Smoke and fire detection

• Line crossing detection

• Camera damage/ overlap/ vandalism detection

• Zone intrusion detection

Detection of loud sounds

This solution uses the video camera built-in microphone to identify sudden and abnormally loud sounds in the surrounding environment. The system analyzes the incoming audio stream in real time, and when predefined threshold values are exceeded, it automatically registers the event and triggers a reaction according to the configured parameters.

Loud sound detection – is a specific Analytics event that is registered by the system when the established sound signal threshold level is reached. Detailed information on how to configure analytics cases and set up automatic reactions to detected events can be found on the Add an Analytics case: Detection of loud sounds documentation page.

Areas of use:

• Industrial site monitoring – detection of emergency sounds and equipment signals.

• Noise level control in residential and public spaces – response to excessive background noise.

• Public safety systems – identification of gunshots, explosions, or screams.

• Integration into smart city infrastructure – enhancing urban safety and responsiveness.

Smoke and fire detection

This is an automated tool designed to locate and detect smoke and open flames within a designated area of the video stream. The system continuously analyzes the video stream in real time to identify signs of smoke or fire. When predefined threshold values are exceeded, the system automatically registers an event and triggers a corresponding response according to the configured parameters.

Smoke and fire detection – is a specific Analytics event that is registered by the system when visual signs of smoke or open flames are identified within the designated area of the video stream. Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Smoke and fire detection documentation page.

Areas of use:

• Monitoring high‑risk fire zones at industrial, commercial, and administrative facilities (production lines, chemical workshops, warehouse complexes).

• Control of small enclosed spaces such as storage rooms, server rooms, and utility areas.

• Safety monitoring in residential and public buildings.

• Automated early fire detection systems within transport infrastructure (subways, railway stations, airports, tunnels).

Line crossing detection

This is an automated tool designed to locate and detect the fact of crossing a designated line in a video stream. The system operates in real time, registering the moment when an object in the frame moves across the designated line in the specified direction. As a result, the system automatically generates an event and activates the predefined response scenario.

Line crossing detection – is a specific event type generated for cases of this Analytics type. Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Line crossing detection documentation page.

Areas of use:

• Access control at secured sites and facilities (warehouses, production workshops, office complexes).

• Perimeter monitoring and surveillance of fenced areas to prevent unauthorized intrusion.

• Automation of security in public and commercial buildings with high traffic.

Camera damage/ overlap/ vandalism detection

A tool for intelligent monitoring of video camera operability designed for automatic detection, localization, and notification of any equipment malfunctions or intentional external impacts. When such events are detected, the system automatically generates an event and activates the predefined response scenario. This tool ensures continuous monitoring of video streams, guaranteeing that each camera consistently performs its primary function without interruption.

The algorithm continuously analyzes the video stream in real time and can detect several critical events:

• Loss of video signal – recognition of connection failure or complete camera shutdown.

• Damage or vandalism detection of attempts to obstruct the field of view (for example, with objects, fabric, or paint), sudden camera rotation, or other actions intended to disable the device.

• Image degradation – identification of severe quality issues such as blurring from defocus or overexposure from bright light sources, which render the video unusable.

Camera damage/ overlap/ vandalism detection – is a specific event type generated for cases of this Analytics type.

Detailed instructions on how to configure Analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Camera damage/ overlap/ vandalism detection documentation page.

Zone intrusion detection

Intelligent video Analytics for automatic motion detection – this tool delivers precise monitoring of object movement within critical surveillance zones. It quickly identifies when a person or object appears or moves in the defined area of the video stream. Once detected, the system automatically generates an event and triggers the appropriate response scenario.

The implementation of Zone intrusion detection Analytics is essential in areas that require precise monitoring of sensitive zones within a site or facility. It enables security teams to focus only on genuine incidents, reducing false alarms caused by movement outside the controlled area.

Zone intrusion detection – is a specific event type generated for cases of this Analytics type. Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Zone intrusion detection documentation page.

Analytics. Level 2

Analytics solutions at Level 2 provide analytics for counting human traffic, with the ability to determine gender, age, and visualize objects within the frame. It also supports dynamic visitor counting (by hours, days, and weeks) to analyze attendance patterns. Level 2 includes:

• People counting in an area

• Visitor counting

People counting in an area

A tool that automatically detects and determines the exact number of people within a designated area of the video stream. When the specified threshold is reached, the system generates an event and activates the predefined response scenario. This solution is widely used in situations that require continuous monitoring of occupancy in specific areas or designated zones.

Counting people in an area – is a specific event type generated for cases of this Analytics type.

Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Counting people in an area documentation page.

Areas of use:

• A highly demanded solution for retail: monitoring occupancy levels in halls, lobbies, and shopping areas.

• An effective tool for tracking attendance and managing visitor flows to optimize service and ensure safety.

• Monitoring of security zones and restricted‑access areas.

Visitor counting

The tool automatically detects and accurately counts people crossing a designated virtual line in the video stream within a specified time period. In addition, the system recognizes the gender and age of visitors. Once detection occurs, it generates an event and triggers the predefined response scenario.

Visitor counting – is a specific event type generated for cases of this Analytics type.

Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Visitor counting documentation page.

Areas of use:

• Analysis of visitor flow and assessment of overall site attendance.

• Research on attendance dynamics of the facility by month, day of the week, hour, and even minute.

Analytics. Level 3

Analytics solutions at Level 3 focus on object recognition and identification: human faces verified against a reference database, vehicle license plates, and cargo container numbers. Level 3 includes:

• Face recognition

• License plate recognition

• Container number recognition

Face recognition

The tool is designed for automatic detection of human faces in a video stream and subsequent identification of individuals based on a reference image (photo or graphic resource) stored in the database. As a result, the system automatically generates an event and triggers the predefined response scenario.

The following specific events are generated for cases of this Analytics type:

• Face recognized – the event is registered when a face captured from the camera's video feed matches a reference image in the database with a similarity score that is equal to or higher than the configured threshold. The required similarity score (threshold) is configured separately for each camera running the Face recognition Analytics.

• Face unrecognized – the event is registered when a face captured from the camera's video feed matches a reference image in the database with a similarity score that is lower than the configured threshold. The required similarity score (threshold) is configured separately for each camera running the Face recognition Analytics.

Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Face recognition documentation page.

Areas of use:

• Integration with smart intercoms – doors can be unlocked automatically when a resident or guest is recognized, eliminating the need for keys or codes. For details, on the Intercom module, refer to the Intercom module documentation section.

• Access control systems – identity verification at entry points to secure facilities such as offices, data centers, or production sites.

• Employee and guest identification – recognition of registered individuals for faster entry and personalized service.

• Unknown face detection – triggers security alerts when unrecognized individuals appear in monitored areas.

License plate recognition

The tool automatically detects vehicle license plates in the video stream and identifies them by matching the alphanumeric sequence against existing database records. As a result, the system automatically generates an event and triggers the predefined response scenario.

The following specific events are generated for cases of this Analytics type:

• Recognized license plate – an event is registered when the license plate image captured from the video stream matches the reference alphanumeric sequence stored in the database.

• Unrecognized license plate – an event is registered when the license plate image captured from the video stream doesn’t match the reference alphanumeric sequence stored in the database.

Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: License plate recognition documentation page.

Areas of use:

• Access control – creation and management of vehicle license plate lists that define which cars are allowed or denied entry to the site (“white” and “black” lists).

• Traffic monitoring – automatic recording of entry and exit times to track working hours or time spent on the premises.

• Logistics – tracking and registering commercial vehicle usage.

Container number recognition

The tool automatically detects and recognizes container numbers within a designated area of the video stream. When recognition occurs, the system generates an event and triggers the predefined response scenario.

Container number recognized – is a specific event type generated for cases of this Analytics type.

Detailed instructions on how to configure analytics cases and set up automatic responses to detected events can be found on the Add an Analytics case: Container number recognition documentation page.

Areas of use:

• Logistics and warehouse infrastructure – automatic registration of container arrivals and departures.

• Customs and control zones – verification of container numbers against declarations.

• Production sites – tracking containers with raw materials or goods.