Visualization of traffic-incident reconstructions with modern JavaScript technologies

OData support
Dr. Forstner Bertalan
Department of Automation and Applied Informatics

Insurance fraud is an enormous problem for car insurance companies, creating a large deficit around the world. Significant amounts are invested in identifying fraud, yet estimations show that half of these still go undetected. A modern solution to this is installing telemetric devices in vehicles, so that a more precise verdict can be reached upon claims announcement. Currently there is no universal solution, insurance companies must build their own platforms if they want to use this data.

The web application came to be as part of a platform which can process this raw dataset, determine if there were any impacts, harsh accelerations or brakings, and connect to other services in order to augment the result with valuable data. The application can then visualize this data on an interactive map, which is given to the claims handlers of compaines wishing to use the platform, they can integrate it into their workflow, so that they get an all-around view of the incident. The incident path and other information is placed on an overhead map, and can be controlled by video-playback controls, making the incident inspectable step-by-step.

In the course of my thesis I will review the tools and technologies I used to create the application. I will also present the architecture of the platform, give a closer look upon the design of the map and the animation, and the finer details of their implementation. Finally I will introduce the tools I have assembled and used to test the application.


Please sign in to download the files of this thesis.