*
AI-powered video analytics in smart cameras with an accuracy of up to


*

AI-powered video analytics in smart cameras with an accuracy of up to

Save on server space and data traffic
Enjoy reduced costs with edge computing
Without any additional steps or human factors
Gain real-time insights
Respond quickly to potential threats
Improve safety
Video analytics in existing smart cameras cannot offer server-level performance. These cameras typically are limited to simple video analytics that are similar across different camera models.
While neural networks can offer solutions for complex business issues, they require significant computer resources.
With our solution, we have optimized neural networks on a single-board computer. This allows us to achieve server-like performance on a compact smart camera.
Smart camera with built-in AI
Корпус смарт-камеры
Artificial Intelligence in smart camera
Tailored to your needs
Flexibility is the important key. A smart camera is assembled as a kit that can handle any task based on an edge device, a compact computing unit.
Various lenses, additional sensors and peripheral modules are available.
Edge-устройство CAMAI
CAMAI
SMART CAMERA
A video analytics system designed from the ground up
Video streams empowered by CAMAI are fully processed on the camera via machine vision and neural networks.
Platform administrators receive data processing results, such as real-time alerts and photo/video of recorded events.
OPTICAL and
IR SENSOR
For biometric identification. This sensor integrates with access control systems for a high level of security.
The IR sensor allows the user to measure body temperature and block access for persons with high temperatures.
Compact
EDGE-DEVICE
Video analytics for existing video surveillance systems
Transforms any IP camera to smart camera.
Offers integration for different manufacturers' cameras into a single video analytics system.
Vandal-proof metal housing
Ethernet, Wi-Fi, LTE
12V, PoE, battery
Easy integration with external systems
Operating temperature range -45 to +50°C
Functional backend for smart camera remote configuration
IP67
IP67
Convenient interface for visualizing data and information
Lenses for different tasks

CAMAI
vs. server-based video analytics

Server-based video analytics
CAMAI
AI onboard
Optimized neural networks
AI smart camera with server-like performance
Optimized hardware
One entry-level server can support thousands of smart cameras
Internet data traffic significant cost reduction. Only alerts and relevant photo/video are transmitted
Data transfer cost reduction
Secure data transfer
Reliable data protection that does not require service engineers with specialized training.
All that is required are new cameras. Only minimum requirements for communication circuits
Limitless scalability
High performance
Expensive server hardware
Neural networks
Expensive communication circuits and data transfer. Video streams from CCTV cameras are continuously transmitted to servers for additional processing
Expensive communication circuits and data transfer. Video streams from CCTV cameras are continuously transmitted to servers for additional processing
Expensive data transfer
Data security issues
Difficult to organize secure video streaming to servers, requiring qualified personnel
Scalability is limited by the capabilities of server hardware and communication circuits
Limited scalability
CAMAI can function even on moving objects, where it would be impossible to run server-based video analytics

CAMAI
vs. server-based video analytics

Server-based video analytics
CAMAI
AI onboard
Optimized neural networks
AI smart camera with server-like performance
Optimized hardware
One entry-level server can support thousands of smart cameras
Internet data traffic significant cost reduction. Only alerts and relevant photo/video are transmitted
Data transfer cost reduction
Secure data transfer
Reliable data protection that does not require service engineers with specialized training.
All that is required are new cameras. Only minimum requirements for communication circuits.
Limitless scalability
High performance
Expensive server hardware
Neural networks
Hard-to-predict costs for powerful GPUs servers
Expensive communication circuits and data transfer. Video streams from CCTV cameras are continuously transmitted to servers for additional processing
Expensive data transfer
Data security issues
Difficult to organize secure video streaming to servers, requiring qualified personnel.
Scalability is limited by the capabilities of server hardware and communication circuits.
Limited scalability
CAMAI can function even on moving objects, where it would be impossible to run server-based video analytics
Usage Scenarios
Transport
Passenger counting for public transport
Attention control for drivers
Safety control for passengers
Vehicle classification
Traffic jam detection
Pedestrian counting at intersections
Smart parking
Industrial video analytics
Personal protective equipment (PPE) control
Personnel vigilance monitoring
Detection for personal cell phone use at work
Restricted area control
Technological regulations and processes control
Employee and contractor monitoring
Smoke and fire detection
Building and construction
PPE control
Employee count
Safety control
Education
Biometric identification
COVID precautions: non-contact body temperature measurement via smart camera
Student vigilance control
Security of premise and perimeter protection
Weapon detection
Warehouse video analytics
Forklift speed measurement
PPE control
Logistics process control
Retail
People counter, as well as gender and age detection
Crowd and queue control
Heat map displaying visitor activity
Biometrics for tracking employee hours
COVID precautions: non-contact body temperature measurement via smart camera
Agriculture
Agricultural crop monitoring
Livestock monitoring
Special-purpose machine monitoring
Bank
Biometrics for personal identification
Automatic scheduling control
Detection for personal cell phone use at work
Crowd and queue control
Didn't find the application scenario you’re looking for?
Contact us about your particular scenario, and we can create a neural network tailored to your needs.
Material control
Biometrics for tracking employee hours
Behavior detection for unruly visitors
Smoke and fire detection
Behavior detection for unruly individuals
Crowd detection
Abandoned item detection
Biometric identification
Smart city
Usage Scenarios
Transport
Passenger counting for public transport
Attention control for drivers
Safety control for passengers
Vehicle classification
Traffic jam detection
Pedestrian counting at intersections
Smart parking
Industrial video analytics
Personal protective equipment (PPE) control
Personnel vigilance monitoring
Detection for personal cell phone use at work
Restricted area control
Technological regulations and processes control
Employee and contractor monitoring
Smoke and fire detection
Building and construction
PPE control
Employee count
Safety control
Education
Biometric identification
COVID precautions: non-contact body temperature measurement via smart camera
Student vigilance control
Security of premise and perimeter protection
Weapon detection
Warehouse video analytics
Forklift speed measurement
PPE control
Logistics process control
Retail
People counter, as well as gender and age detection
Crowd and queue control
Heat map displaying visitor activity
Biometrics for tracking employee hours
COVID precautions: non-contact body temperature measurement via smart camera
Agriculture
Agricultural crop monitoring
Livestock monitoring
Special-purpose machine monitoring
Bank
Biometrics for personal identification
Automatic scheduling control
Detection for personal cell phone use at work
Crowd and queue control
Didn't find the application scenario you’re looking for?
Contact us about your particular scenario, and we can create a neural network tailored to your needs.
Material control
Biometrics for tracking employee hours
Behavior detection for unruly visitors
Biometric identification
Abandoned item detection
Behavior detection for unruly individuals
Crowd detection
Smoke and fire detection
Smart city
Cases
Result:
Task:
Labor safety.
Employee identification.
Employee vigilance control.
Technology regulation compliance.
Number of employee violations decreased by 72% in a month after platform implementation.
Number of violations related to employees’ not following safety protocols decreased by 75%.
Number of restricted area trespassing incidents decreased by 92%.
Employee count.
Implemented scenarios:
Employee scheduling control.
Personnel vigilance monitoring and detection for personal cell phone use at work.
Biometrics and restricted area control.
Technology regulation control.
PPE control.
Cases
Task:
Incoming and outgoing passenger counting in real-time with regular bus cameras.
Optimization of ticket inspector costs
Reduction of passengers travelling without tickets.
Passenger counting. Integration with bus onboard computer to compare the passengers with the number of travel document validations. Calculated potential number of free riders increases the ticket inspector’s efficiency.
Implemented scenarios:
Result:
Counting accuracy of more than 96% was achieved using regular bus cameras.
Cases
Result:
Task:
Ensure safety for a beverage manufacturer’s warehouses. Respond promptly to violations and emergency situations.
Number of employee violations decreased by 87% in a month after platform implementation. PPE-related violations decreased by 97%.
Forklift speed measurement and dangerous maneuver control.
Implemented scenarios:
Restricted area control with dynamic boundary adjustments (for cargo handling operations).
PPE control.
Cases
Result:
Task:
Labor safety.
Employee identification.
Employee vigilance control.
Technology regulation compliance.
Number of employee violations decreased by 72% in a month after platform implementation.
Number of violations related to employees’ not following safety protocols decreased by 75%.
Number of restricted area trespassing incidents decreased by 92%.
Employee count.
Implemented scenarios:
Employee scheduling control.
Personnel vigilance monitoring and detection for personal cell phone use at work.
Biometrics and restricted area control.
Technology regulation control.
PPE control.
Cases
Task:
Incoming and outgoing passenger counting in real-time with regular bus cameras.
Optimization of ticket inspector costs.
Reduction of passengers travelling without tickets.
Passenger counting. Integration with bus onboard computer to compare the passengers with the number of travel document validations. Calculated potential number of free riders increases the ticket inspector’s efficiency.
Implemented scenarios:
Result:
Counting accuracy of more than 96% was achieved using regular bus cameras.
Cases
Result:
Task:
Ensure safety for a beverage manufacturer’s warehouses. Respond promptly to violations and emergency situations.
Number of employee violations decreased by 87% in a month after platform implementation. PPE-related violations decreased by 97%.
Forklift speed measurement and dangerous maneuver control.
Implemented scenarios:
Restricted area control with dynamic boundary adjustments (for cargo handling operations).
PPE control.
Project approval
Neural network model customization
Environment survey
Additional training for the neural network
03
02
04
Defining Key Performance Indicators.
Achieving target accuracy in the laboratory.
We are open to collaborating with system integrators
Videorecording and data collecting to enrich datasets.
Adjusting the neural network for the real environment.
01
Smart camera customization
Smart camera production
Software porting
Equipment installation
07
06
08
Electronic component selection and housing adaptation for future operating conditions.
Installing equipment at the customer's premises.
Software adaptation to customized smart cameras.
Assembling cameras with the necessary set of sensors and network interfaces.
05
Test operation
Project handover
Additional training for the neural network
Technical support
11
10
12
Test system launching.
Additional cycles of neural network training to increase accuracy (if necessary).
Acceptance tests, demonstrating project indicators have been achieved. Commercial operation.
09
Project plan
for large-scale projects
4-6 months
AB Technology supports projects at all stages.
Project approval
Smart camera customization
Test operation
Neural network model customization
Smart camera production
Project handover
Environment survey
Software porting
Additional training for the neural network
Additional training for the neural network
Equipment installation
Technical support
03
07
11
02
06
10
01
05
09
04
08
12
Defining Key Performance Indicators.
Electronic component selection and housing adaptation for future operating conditions.
Test system launching.
Videorecording and data collecting to enrich datasets.
Software adaptation to customized smart cameras.
Additional cycles of neural network training to increase accuracy (if necessary).
Adjusting the neural network for the real environment.
Assembling cameras with the necessary set of sensors and network interfaces.
Acceptance tests, demonstrating project indicators have been achieved. Commercial operation.
Achieving target accuracy in the laboratory.
Installing equipment at the customer's premises.
AB Technology supports projects at all stages.
for large-scale projects
4-6 months
Project plan
We are open to collaborating with system integrators

CAMAI technology

Accelerate data processing on a single board computer, with speeds up to 30 to 50 times faster compared to running unoptimized server code.

Acquiring video stream
Applying computer vision methods
Neural network optimization
Operations minimization
Parallel
processing
Established in 2004, we are an IT company that provides innovative AI-powered video analytics software solutions for improved security and safety for governments and organizations.
AB Technology
Transport Innovations of Moscow accelerator finalist.
Tech Explorer Accelerator 2021 winner.
Skolkovo innovation center resident.
highly qualified employees, including experienced developers, programmers, data analysts, and industrial designers.
Trusted by
Video analytics
at the edge
for your needs
Reduce costs associated with hardware and communication circuits

Unique solution for geographically separated video analytics systems

64 Lesnoy Avenue,
Saint-Petersburg,
Russia
*According to Biometrics testing results on independent databases LFW, Feret, MegaFace, etc. (3 million photos).
Copyright © 2022 AB Technology All rights reserved
64 Lesnoy Avenue,
Saint-Petersburg,
Russia
*According to Biometrics testing results on independent databases LFW, Feret, MegaFace, etc. (3 million photos).
Copyright © 2022 AB Technology All rights reserved