DIMOP_PLUSZ-1.1.2/B-24-2025-00088
Adaptive object recognition and positioning from a moving source
Project data
- Name of the beneficiary
- UDEMX Szoftverfejlesztő és Tanácsadó Korlátolt Felelősségű Társaság (UDEMX Software Development and Consulting Ltd.)
- Project identification number
- DIMOP_PLUSZ-1.1.2/B-24-2025-00088
- Subject of the development (project title)
- Adaptive object recognition and positioning from a moving source
- Amount of the contracted grant
- HUF 79,545,150
- Total eligible cost of the project
- HUF 119,634,500
- Grant intensity
- 66.49%
- Start of implementation
- 1 March 2026
- Planned completion date
- 31 October 2027
- Place of implementation
- 6725 Szeged, Móra utca 49. Fsz. 1. ajtó, Hungary
- Programme
- Digital Renewal Operational Programme Plus (DIMOP Plusz)
- Call for proposals
- Dedicated R&D&I and market-entry programme for actors of the digital economy – Corporate digital R&D developments (DIMOP_PLUSZ-1.1.2/B-24)
- Managing authority
- Ministry of Energy of Hungary
- Source of funding
- European Regional Development Fund (ERDF) and the central budget of Hungary
About the project
Under the Digital Renewal Operational Programme Plus, UDEMX Ltd. has been awarded a non-repayable grant to research and develop software technology that recognises objects in real time from a video stream captured by a moving source – for example a camera mounted on a vehicle or another moving device – and determines the spatial position of the objects it identifies.
The difficulty of the task comes from the fact that the recording device itself is in motion: the scene changes continuously, the viewing angle, the lighting and the sharpness of the image are unpredictable, and the data used for positioning arrives with varying accuracy and varying timestamps. The aim of the project is to develop an adaptive method that adjusts to these changing conditions on the fly, and that combines image-based recognition with positioning data to produce a single, reliable result.
Implementation is organised into four phases:
- Literature review, evaluation and specification. Collecting, analysing and evaluating the scientific literature on image processing and the related artificial intelligence algorithms, and examining the hardware devices relevant to the development from a software engineering perspective. Output: a summary research document and a development specification.
- Modelling. Algorithm modelling based on the specification, with particular attention to linking data arriving from a moving source with data supporting positioning. Output: documentation summarising the modelling research.
- Development of the data recording software tool. Implementation of the software tool based on the specification and the results of the modelling phase, extraction of sample data matching the data communication of the various hardware devices, development of the image recognition and positioning algorithms, compilation of a training database and training of artificial intelligence algorithms. Output: complete software with accompanying documentation.
- Integration, testing and validation. Integration of the completed components, testing and operational validation. Output: an integrated, tested and validated software package with test records and accompanying documentation.
The project comprises both industrial research and experimental development activities, and is carried out in Szeged with the involvement of six UDEMX specialists. The results of the development complement UDEMX's software development and consulting services and will broaden the company's market offering.
The software is being developed with the requirements of digital accessibility in mind.
A projekt az Európai Unió támogatásával, a magyar állam társfinanszírozásával valósul meg.
The project is implemented with the support of the European Union and co-financing by the Hungarian State.
