EN
VN

Smart Security

Creating a safe and secure environment is essential for the success and sustainability of any buildings and cities. It is, however, often challenged by the lack of manpower, integration, ongoing updates and maintainance. Our AI technology promise to deliver a comprehensive security solution that is effective, cost-saving and easy to implement for any residential buildings, schools, hospitals, urban cities, hotels or commercial centers.

Our AI Solution

Real-time detection of abnormal behaviors helps avoid the risk of terrorism, rescue people in danger, contributing to the maintenance of a civilized lifestyle. In order to enhance smart security in public areas, we have built a comprehensive AI system that can automatically detect abnormal behaviors through surveillance cameras.  

Currently, our smart security system can detect 09 types of abnormal behaviors. Some behaviors are sophisticated or very sophisticated such as Kidnapping, Harassment. 

 

  • Kidnapping Detection 
  • Harassment Detection 
  • Fighting Detection 
  • Weapon Detection 
  • Drunk people Detection 
  • Lying down Detection 
  • Upper Half Naked People Detection 
  • Pet (dog and cat) Detection 
  • Climbing Detection 

 

Climbing Detection

Drunk People Detection

Fighting Detection

Harrassment Detection

Kidnapping Detection

Lying Down Detection

Pet (Dog and Cat) Detection

Upper Half Naked People Detection

Weapon Detection

Product Highlights

  • AI 24/7 automatic monitoring and alerting service for abnormal behaviors help building managers to quickly resolve residents' pain points & rescue residents from danger promptly.
  • Deep Learning techniques enable the robustness of our aberrant behavior detector under the diversity of the environments and camera locations.
  • Easy to control & maintain on a thousand of CCTV cameras via our complete solution with capability of deploying AI models on both edge-device and cloud.

With the aim to enhance security in public areas, we developed the face recognition system that automatically recognizes the people of interest from CCTV videos, i.e., the citizen, the blacklist, or the stranger. This system has been deployed at OceanPark, an urban area of Vinhomes.

Face Recognition

Product Highlights

  • Our face recognition model employs deep learning technique and is trained on a million-scale and diverse dataset.
  • It is able to recognize people who appeared on the surveillance cameras at different lighting conditions, nearly reach the human visual ability with an F1 of more than 90%.
  • The face recognition system has been built to stream the CCTV data directly and return the detection results in real-time.

We developed an AI solution for searching the time and location of an item in a video given a photo or estimated lost time of the item. In the area with CCTVs, therefore, building managers can have a better property control & support their residents in retrieval of the lost item by applying our solution. 

Lost-item retrieval

Product Highlights

  • Retrieving the objects appeared on the CCTV within 5 meters at different lighting conditions with more than 89% accuracy as our model employed state-of-the-art object detection and metric learning techniques.
  • Recapturing best on three types of objects, i.e. handbag, backpack, and suitcase and more object types in the future e.g. handphone and wallet.

Along with the increasingly large-scale vehicles, the automatic traffic monitoring system including camera and speedometer is widely distributed throughout the metropolitan area. We innovated AI solutions that can recognize & alert the most popular scenarios in the traffic eco-system to help smooth traffic inside and outside the residential area: 

 

  • Traffic Jam Detection 
  • Obstacle Detection 
  • Parking Violation 

 

Traffic monitoring

Product Highlights

  • Automatic monitoring of real-time traffic footage, support 10 CCTV cameras at the same time, work on a variety of lighting and weather conditions.
  • Early detecting obstacles & alerting observer of the building to remove them beforehand, avoid traffic jam & danger for drivers.
  • Recognizing & warning with traffic jam, parking violation

Usage Scenarios

Urban areas

Schools

Hotels/ Resorts

Hospitals

Commercial Centers

Factories

Top

Share