Abstract

ADDPRIV proposes novel knowledge and developments to limit the storage of unnecessary data, to be implemented on existing multicamera networks in order to make them better comply with citizen´s privacy rights. ADDPRIV tackles the challenge of determining in a precise and reliable manner private data from video surveillance which is not relevant from the perspective of security and which need not to be stored. ADDPRIV proposes solutions for automatic discrimination of relevant data recorded on a multicamera network, related to an individual whose suspicious behaviour triggered an alert. Relevant data not only corresponds to video scenes capturing individuals´s suspicious behavior (smart video surveillance), but also automatically extracting images on these individuals recorded before and after the suspicious event and across the surveillance network. Algorithms for automatic browsing, identification and retrieval of images on the suspicious individuals across the multicamera network will be developed, implemented and validated in a real life scenario. These solutions will be able to discriminate data relevancy and index relevant images, setting different storage levels with different data access management. Solutions for secure deletion of non-relevant video files will also be developed, accordingto the highest security standards. ADDPRIV will therefore provide required tools for limiting the visualization and storage of unnecessary data, corresponding to law abiding citizens. ADDPRIV solutions´ validation will be guided by criteria and metrics determined by social and ethical experts and end users, establishing the characteristics that a surveillance system must fulfill for effectiveness and integrity in citizens privacy protection. ADDPRIV will analyse the impact of our system on citizen´s human rights and end users organizational processes, aiming to contribute towards establishing a standard on how and when these systems are to be used.