Predictive modelling of real time video of outdoor scenes captured with a moving handheld camera

Purpose and goal

The goal of the project has been to create a functional prototype of an innovative software for robust predictive modelling of real time video of outdoor scenes captured with a moving camera. The project has reached this goal and developed methods are now being integrated in several of the products of the company. Through an extensive R&D effort, algorithms that are able to track and trace sub-pixel sized objects moving at speed within a moving video stream have been development. The collaboration with the academic part has, in addition, led to the publication of two scientific articles.

Expected results and effects

The project has enabled Protracer to further strengthen its growth potential and ability to innovate. With the new model, Protracer has broken new ground on the international market and reached long-term competitive advantages. The project has generated several spin off results which have found use in other products. Researchers at Uppsala University are currently evaluating utilization of developed algorithms for a range of biomedical applications.

Planned approach and implementation

The projects was divided into four work packages. The initial market analysis clearly demonstrated the need. The scientific analysis identified a combination of suitable methods, to test and further develop. These methods were integrated into a prototype. The following evaluation highlighted pros and cons of the selected approach. Development phases were iterated with thorough testing in order to reach the desired functionality and reliability. The project has been realized in collaboration with the Centre for Image Analysis, Uppsala University.