KIRAS Security Research

Call results by calendar year

Call results by funding instruments

Projects of the joint German-Austrian call

Cooperative Research and Innovation Projects >Call 2018 >

MaLeStV

Machine Learning of motion patterns in the penal system

The everyday life of a prison is characterised by aspects of security. A number of structural and organisational measures (prison cell design, video surveillance, ...) are designed to reduce the risk of safety-related incidents in prisons and to support the personnel of the correctional system in their work. However, a permanent observation of potential endangering and endangered persons is not feasible due to limited personnel resources. Violence in prison is no exception: suicides, for example, can be prepared and carried out in unobserved moments. Suicides in prisons often cause collateral damage beyond the tragedy of the event, ranging from increased media coverage to irritation among employees and inmates. This is where the proposed project comes in. With the help of new technologies of 3D image analysis, behavioural patterns of people are registered with the help of a 3D sensor and analysed in real time in order to identify corresponding critical movement patterns. Based on this real-time analysis, an alarm can be triggered to alert the security personnel to the potential danger situation so that intervention can take place at an early stage. Since this technology does not use video images, but is based on three-dimensional sensor data of motion sequences, a number of data protection problems that have to be taken into account when using video technology are omitted. The integration of such an innovative, sensor-based system in the prison leads to an optimisation of the justice system's performance: this is demonstrated by more security in the prison, from a preventive point of view the risk of illegal and criminal behaviour can be reduced, and routine activities such as the live surveillance of a prison cell can be made more effective. In the longer term, the analysis quality of such a system can also be improved through the use of adaptive software. This can reduce the rate of false positive alarms and thus increase the effectiveness of the measure. 

Project leader
DI Michael Brandstötter, MSc(OU)
Cogvis gmbh
Wiedner Hauptstrasse 17/3a, 1040 Wien
+43 1 2360580
brandstoetter@cogvis.at
https://cvl.tuwien.ac.at/project/malestv/ 

ProjektleiterIn
Technische Universität Wien, Computer Vision Lab
Institut für Höhere Studien
Bundesministerium für Verfassung, Reformen, Deregulierung und Justiz