Oddity: Increasing public safety with help from Oddity’s algorithm

Published on: 22 November 2020

The Dutch camera observation landscape within the public domain currently has a reactive set-up. Observers receive alerts from police units on location to observe and monitor via the camera systems. With Oddity's algorithm there is a shift to a proactive working method, in which the system warns the observer when deviating situations occur.

This allows operators to work based on triggers, which increases efficiency and focus. In addition, there is an increased chance that important incidents will be noticed in time by the observer. Currently, Oddity is already making several cities safer by signaling violence. In the future, it is expected that all activities that can be recognized by an observer can also be recognized by the algorithm.

Deployment of Artificial Intelligence

The application of deep learning technology in camera surveillance is still in its infancy. The underlying techniques have only recently become economically attractive to apply to video images, because they require relatively large amounts of computer power. Oddity is one of the first Dutch parties to be able to use violence recognition to improve public safety. Solutions such as Video Content Analysis (VCA) use traditional algorithms to add limited functionality to camera surveillance. Think of people counting, line crossing and lingering detection; no techniques based on deep learning are used. Oddity’s algorithm uses the latest techniques to analyze video images with relatively little computer power.

What challenge does it solve?

Oddity solves a number of problems in camera surveillance:

  • No more violent incidents are missed in the surveillance room
  • It ensures fast detection and response times
  • Operators work in a focused and more efficient way
  • Surveillance is improved without adding personnel

Oddity participated in the SBIR “Artificial intelligence for public services”. Through the cooperation with the NL AIC, certain (semi-)governmental parties become more accesible. With the budget of the SBIR and the platform provided by the coalition, Oddity’s position in the surveillance areas of the Netherlands will be strengthened.

Small Business Innovation Research (SBIR)

SBIR makes use of an exception in the procurement legislation for applied research and development. A societal question posed by the government is central.  SBIR works in a tiered innovation competition towards the innovative solution(s). The best proposals execute a feasibility study and the best feasibility studies are commissioned to develop an innovation. This innovation is ideally tested in practice by the potential government customer.

In collaboration with:

Oddity was created by three enthusiastic entrepreneurs whose mission is to raise public safety to a higher level by means of machine learning. Together with top researchers, Dr. R.W. Poppe and Dr. R.L. Jansen, within the University of Utrecht and with the help of coaches and experts, they have built up many contacts within the world of camera surveillance over the past year.

More information

Interested? Visit the website of Oddity for more information.

Share via: