Research Article
A System for Privacy-Preserving Analysis of Vehicle Movements
@INPROCEEDINGS{10.1007/978-3-319-67636-4_3, author={Gianluca Lax and Francesco Buccafurri and Serena Nicolazzo and Antonino Nocera and Filippo Ermidio}, title={A System for Privacy-Preserving Analysis of Vehicle Movements}, proceedings={Cloud Infrastructures, Services, and IoT Systems for Smart Cities. Second EAI International Conference, IISSC 2017 and CN4IoT 2017, Brindisi, Italy, April 20--21, 2017, Proceedings}, proceedings_a={IISSC \& CN4IOT}, year={2017}, month={11}, keywords={Privacy Vehicle movements Beaglebone JavaANPR}, doi={10.1007/978-3-319-67636-4_3} }- Gianluca Lax
Francesco Buccafurri
Serena Nicolazzo
Antonino Nocera
Filippo Ermidio
Year: 2017
A System for Privacy-Preserving Analysis of Vehicle Movements
IISSC & CN4IOT
Springer
DOI: 10.1007/978-3-319-67636-4_3
Abstract
In this paper, we deal with the problem of acquiring statistics on the movements of vehicles in a given environment yet preserving the identity of drivers involved. To do this, we have designed a system based on an embedded board, namely Beaglebone Black, equipped with a Logitech C920 webcam with H.256 hardware encoder. The system uses JavaANPR to acquire snapshots of cars and recognize license plates. Acquired plate numbers are anonymized by the use of hash functions to obtain plate digests, and the use of a salt prevents plate number discovery from its digest (by dictionary or brute force attacks). A recovery algorithm is also run to correct possible errors in plate number recognition. Finally, these anonymized data are used to extract several statistics, such as the time of permanence of a vehicle in the environment.
