Ensuring security and order in Austrian prisons is a core task of the Austrian penal system. In particular, the protection of persons directly involved in the prison system, i.e. inmates and staff, against physical and psychological violence is of essential importance. In 2019, there was an average of one physical assault on staff in Austrian prisons per day and more than two reported criminal acts between prisoners per day.
The aim of the project is to investigate how new technologies from the field of artificial intelligence can be used to reduce the workload of prison staff and better protect both staff and detainees. To this end, an innovative multi-modal approach is being taken, applying insights from the emerging research area of “Privacy Preserving Machine Learning” (PPML). Already during the acquisition of sensor data, privacy preservation is ensured by increasingly relying on anonymizing sensor technologies. Specifically, 3D and thermal sensors, as well as "wearables" will be used; conventional RGB cameras will only be activated context-sensitively if the identification of persons is unavoidable for security reasons. The designed system is open; future extension by other modalities (such as audio or any IoT devices) is also enabled. The different data sources are merged into a consistent fusion model to capture complex behavioral patterns. In addition to critical - for example, health-related - events that require immediate intervention, other events that appear unremarkable when viewed in isolation will also be captured by the system. These include non-verbal interaction or physical contacts. These are stored in an event graph and thus allow the analysis of (among other things, aggression-related) behavior patterns even over longer periods of time. Furthermore, this approach establishes a connection to the concept of "Explainable Artificial Intelligence (XAI)", which is gaining increasing importance, and enables the development of an overall system that presents not only useful but also comprehensible conclusions to the end user. In addition, the technology in question will be subjected to a comprehensive impact assessment. This includes an empirical needs and risk analysis as well as a legal examination and critical ethical discussion of the admissibility of such applications under fundamental and human rights law.
Projektpartner
Projektleitung:
Computer Vision Lab, Technische Universität Wien, Institut 193/1
Ansprechperson: PD. Dr. Martin Kampel
Telefon: 01 58801-193164
E-Mail: martin.kampel@tuwien.ac.at
Adresse: Favoritenstraße 9, 1040 Wien
Homepage: https://cvl.tuwien.ac.at/
Bundesministerium für Justiz
Ansprechperson: Dipl.-Ing. Wolfgang Schlapschy
Telefon: 0650 / 90 20 606
E-Mail: wolfgang.schlapschy@bmj.gv.at
Adresse: Museumstraße 7, 1070 Wien
Homepage: https://www.bmj.gv.at/
CogVis Software und Consulting GmbH
Ansprechperson: Michael Brandstötter
Telefon: 01 236 0580
E-Mail: brandstoetter@cogvis.at
Adresse: Wiedner Hauptstraße 17/3a, 1040 Wien
Homepage: https://cogvis.ai/
PKE Holding AG
Ansprechperson: DI Werner Kloihofer, MLBT
Telefon: +43 664 80470 1486
E-Mail: w.kloihofer@pke.at
Ansprechperson: Patrick Leitgeb, MSc BSc
Telefon: +43 664 80470 1271
E-Mail: p.leitgeb@pke.at
Ansprechperson: Walter Kuhn MAS
Telefon: +43 664 80470 1108
E-Mail: w.kuhn@pke.at
Adresse: Computerstraße 6, 1100 Wien
Homepage: https://www.pke.at/
Research Institute AG & Co KG
Ansprechperson: Robert Rothmann, MA PhD
Telefon: +43/1-524-3524-0
E-Mail: robert.rothmann@researchinstitute.at
Adresse: Wollzeile 1/1/1, 1010 Wien (RI Büro c/o Österreichische Computergesellschaft OCG)
Homepage: https://researchinstitute.at/
Projekthomepage
https://cvl.tuwien.ac.at/project/kiis/
