top of page

Project Optimus Object Detection

Network Rail tasked Synoptix with creating an auto-census device capable of performing on-the-edge object detection in-situ to generate on-site censuses at level crossings. The aim was to provide an automated alternative to managers standing at level crossings with a clipboard, manually counting each type of user.


Our approach

Synoptix collaborated with the University of Leicester for the successful delivery of the project, which included a number of successful site trials and a study to assess the performance benefits of A.I. versus human.

Benefits

Project Optimus is capable of running 24/7, 365 days a year without intervention. It therefore removes the human factors often experienced in census work, such as loss of focus after an extended period of time.


Our solution provides a more time effective and cost efficient method as the data no longer has to be post-processed by external contractors to provide an accurate census. This solution overcomes any issues with human influence affecting results.


Optimus’ on-the-edge solution analyses the data on site rather than providing a cloud-based solution, which removes the need for additional infrastructure. By placing the data analysis on the device, this system can be used in remote locations with minimal infrastructure or human interaction. Because lag due to data transmission is removed completely, the speed from detection to analysis is greatly increased.


Project Optimus was originally designed for the rail network to save time and resources used for census gathering. Now, this technology can be used across multiple sectors – from traffic monitoring on the roads to policing to security applications where accurate detection is critical.

New Project (7).jpg

Project Optimus Object Detection

iStock-1403105138.jpg
Interested in how our systems engineers can help deliver your project on brief, on time and on budget? Contact us to discuss your needs or find out more about how we could work together.
bottom of page