Wagner, P.P.WagnerBirnstill, PascalPascalBirnstillKrempel, ErikErikKrempelBretthauer, S.S.BretthauerBeyerer, JürgenJürgenBeyerer2022-03-132022-03-132017https://publica.fraunhofer.de/handle/publica/39864010.1007/978-3-319-67816-0_11Dashcams are small, dashboard mounted camera systems that continuously monitor the area around a vehicle and record video images on a portable storage device. According to many data protection authorities, dashcams constitute surveillance systems that are operated by private individuals in public places. By continuously acquiring personal data they interfere disproportionately with the right of informational self-determination. One approach to make dashcams compliant to data protection law is to automatically identify personal information - at least pedestrian's faces and license plates - in the captured video image and subsequently disguise them. Even though appropriate anonymization methods exist, high computational costs prevent their use in portable dashcams. This article presents a new approach that enforces the anonymization of encrypted dashcam videos on a dedicated computer system, before the user gets access to the videos. To accomplish this, classified images are safeguarded by usage control techniques on the way from the camera to the anonymization component. By applying the developed system, any existing dashcam can ultimately be enhanced by privacy protection capabilities.enconference paper