Virtus AI

PUBLIC SAFETY SOLUTION

Take immediate action

thanks to the prediction of a crime.

Predict, prioritize and see through the eyes of citizens.

Predict through the use of Machine Learning the probability of the occurrence of a crime, in a particular place, on a particular day and in a time range.

Real-time video understanding can identify events of interest and prioritize operator attention.

 

Connect citizens with safety forces to crowdsource area coverage. ​

Security Systems need to be able to optimize their processes in order to have cataloged and identified the crimes.

It would be great to be able to predict crimes to create strategic police deployment.

Crime Prediction

 

CHALLENGES

•Lack of strategic indicators to analyze crime patterns.

•Predict crimes to reduce insecurity.

•Deploy strategically the police force.

SOLUTION

•  Identification and Segmentation of the zones according to their level of risk.

•  Prediction model to identify the occurrence probability of a crime.

Success Story

Real-time video understanding to prioritize operator attention

 

Despite tremendous costs and potential, these resources are greatly underutilized​

Without Cognitive Video Analytics

  • Human monitoring of live video is very costly and inefficient

  • Human attention limited to 4-9 displays

  • 4 displays = 80% and 9 displays = 40%

  • Operator fatigue (<5% events after 20 minutes)

  • Less than 1% of video is ever seen by human eyes

 

With Cognitive Video Analytics

  • System prioritizes operator attention

  • Only ‘likely events’ need to be confirmed

  • Event type and location are highlighted on the image

  • # of streams per operator limited by false-alert rates

  • E.g., 1,200 streams @ 1 false alert/24 hours

 

Example: 

application OF occluded faCe 

detectioN

Citizens’ Networks

 

Demo

Through the eyes of Citizens

Through a mobile app (i.e. Kaizala) citizens use action cards to report potential crimes.

Feedback from citizens alert police forces.

Unreported crimes now get reported.  (i.e. Statistics & Analytics)

© 2019 by analytikus, LLC - Privacy Policy

us_1821_499x335.jpg
EN
España.png
ES
  • LinkedIn Social Icon
  • Twitter Classic
  • Facebook Social Icon
  • YouTube Social  Icon
  • itunes