QazSmartVision.AI

Intelligent video analytics and computer vision

Company

"QazSmartVision.AI" LLP was founded in 2021 and, during its effective and fruitful tenure in the IT market, has established itself as a qualified, reliable, and efficient partner. We are a systems integrator and provider of telecommunications and digital solutions.
Mission, Vision, Values
1
Mission
We provide effective IT solutions to enhance organizations' productivity, aiming for high-quality, prompt, and transparent fulfillment of end consumers' interests.
2
Vision
To become one of the top 5 best IT companies in Kazakhstan, with high added value, an expanded international franchise network, and leading startups.
Core Values
1
Leadership
We are ambitious in our goals, striving for market leadership and the creation of new markets. Each one of us is a leader and aims to contribute more to the company.
2
Professionalism
We are a team of highly engaged professionals focused on results. We are perfectionists in our work, always aiming for the best solutions, exceeding customer expectations.
3
Reliability
We value our reputation and always seek to justify the trust of our clients, partners, and colleagues. We are open and honest in our relationships with clients, partners, and colleagues.
4
Freedom
We base our relationships on values when choosing clients, partners, and colleagues. We build relationships on trust, delegate authority, and reinforce autonomy.
Key areas of activity
  • Intelligent video analytics and computer vision
  • Development and maintenance of information systems
  • Supply, configuration, and maintenance of server and network equipment
  • Mobile application development
  • Information security and IT audit
Intelligent Video Analytics and Computer Vision
AI Platform is a unique development in the field of computer vision and object recognition of various complexity levels. It ensures efficient and accurate processing of objects in images and video streams, capable of functioning on diverse operating systems and platforms

Facial Recognition

  • Recognition of a face and its properties
    Detection of facial areas in the image and parameters: gaze direction, head tilt, eye status, emotion detection, presence of a mask, glasses, or smile.
  • Facial attributes Recognition
    The system can determine gender and age from the image and can accurately identify adulthood.
  • High accuracy in recognition
    Recognition of a person from a facial image with high precision is confirmed by the System's algorithms applied in numerous projects in various fields.
  • Wide application of facial recognition
    Facial recognition algorithms are used for access control in public places, transportation fare payment, searching for individuals (including missing children), and more.

Intellectual Search

  • Face Detection and Tracking
    The system detects a face and tracks its movement across the network of surveillance cameras. Only a few frames are required for the system to identify a face during inter-camera tracking.
  • Face Search
    Searching for a specific face within the system's recorded events archive takes less than 1 second.
  • Various Search Criteria
    Available search criteria include movement within an area, crossing lines in a chosen direction, prolonged presence of an object in an area, transition from one area to another.
  • State Numbers Search
    The system enables locating where a vehicle was recently observed and in which direction it was moving.

Recognition of Personal Protective Equipment (PPE) and Uniforms

  • Different Types of PPE
    Detectors operate in real-time and allow the identification of various types of personal protective equipment (PPE): helmets, masks, protective clothing.
  • Control of Unauthorized Uniform Wearing
    The system monitors the wearing of uniforms with insignia exclusively by employees by comparing the recognized individual in the uniform with internal databases of individuals, police databases, etc.
  • Monitoring PPE Wearing
    Within 3 seconds, the system can recognize the presence of protective equipment and notify the user of safety rule violations.

Recognition of Associations of Face and Body

  • Facial and Body Recognition
    The system enables simultaneous identification of the facial area and the body area of a person in an image.
  • Association Building
    The system establishes a connection between the detected face and body in the image, allowing the determination of the belongingness of the found face and body to a single individual.
  • Wide Capabilities
    Establishing connections between faces and bodies detected at different time intervals. The system operates effectively in crowded areas with various positions of a person's face and body.

Crowd Recognition

  • Monitoring Crowd Gatherings
    The system performs people counting within frames based on silhouettes and headcounts, notifying the user when a crowd forms within 5 seconds.
  • Activity Monitoring
    The system allows tracking gatherings of people based on several parameters:

    · By time of day and days of the week,
    · By locations of the gatherings.
  • Customizable Response Threshold
    The ability to set the number of people, upon reaching which the System will send a notification for decision-making at the observation point.

Recognition of Fights

  • Pose Recognition during a Fight

    The system enables the recognition of:
    · Raised arms and legs,
    · Active movements of arms and legs,
    · Prolonged crossing of detection zones.
  • Wide Range of Applications

    The system can be utilized to prevent fights:
    · on the streets,
    · in courtyards,
    · in educational institutions,
    at stadiums and arenas during sports and other mass events.

Recognition of Human Poses and Actions

Extensive Pose Recognition
The system allows detection of:
  • A person lying down,
  • Shooting position,
  • Raised arms,
  • Hand placement on a handrail, and more.
Search Using Specific Gestures
The technology identifies instances where assistance is required – when a person raises both arms up, the System sends a notification to the user.

Sudden Body Position Changes

Detecting and notifying about sudden changes in body position allows for detecting falls and other abrupt body movements.
Multi-camera Tracking
  • Movement Tracking

    The technology allows tracking the movements of people on the ground using a network of video cameras without facial identification.
  • Synchronization and Accuracy

    Determining people's locations with up to 96% accuracy in real-time and visualizing processed data.
  • Blind Spots

    Reidentification technology enables finding a person even after they enter a blind spot and appear on another surveillance camera.
  • Identified Attributes
    During tracking, the System identifies the following attributes: gender, clothing color, headwear, sleeve length, presence of a backpack.

Vehicle Recognition

  • Wide Range of Capabilities

    The system recognizes vehicle state numbers, vehicle types, colors, make, and model.
    Functionality expansion is possible through neural network retraining.
  • Various State Number Types

    The system recognizes vehicle state numbers from over 20 countries, including Kazakhstan, Russia, Belarus, EU countries, CIS, etc. The list is continuously updated through neural network retraining.
  • Recognition of Special Vehicles

    Identification of emergency service vehicles (police cars, ambulances, emergency services, etc.)
  • Extensive List of Vehicle Brands and Models
    The system recognizes over 150 vehicle brands and more than 850 models.

Firearm Recognition

  • Weapon Detector

    Triggers upon detecting a firearm in a person's hands in the frame, identifying the "shooter's pose."
  • Use Case Scenario

    The primary use case scenario is public safety (streets, offices, educational institutions, etc.). The system will detect the weapon and send a notification to the operator for response and assistance.
Additionally

BPM Platform
· A unique development from Kazakhstan
· Document signing using the EDS (Electronic Digital Signature) of the National Certification Authority of the Republic of Kazakhstan
· Capability of integration with State Information Systems
Includes the following modules:
1. Business Process Automation:
Complete transition of providing public services into electronic format
2. Geoinformation Systems:
Creation of geospatial engineering projects
Geospatial data bank
3D digital terrain models
Cartographic products and services
Our Partners
Сontacts

ТОО «QazSmartVision.AI»

Republic of Kazakhstan, Astana city, Almaty district, Tauelsizdik Avenue, location 32, 6th floor.


info@qazsmartvision.ai

www.qazsmartvision.ai


+ 7 (7172) 57 00 27

Contact Us