A general framework for managing and processing live video data with privacy protection

Abstract

Though a large body of existing work on video surveillance focuses on image and video processing techniques, few address the usability of such systems, and in particular privacy issues. This study fuses concepts from stream processing and content-based image retrieval to construct a privacy-preserving framework for rapid development and deployment of video surveillance applications. Privacy policies, instantiated to as privacy filters, may be applied both granularly and hierarchically. Privacy filters are granular as they are applicable to specific objects appearing in the video streams. They are hierarchal because they can be specified at specific objects in the framework (e.g., users, cameras) and are combined such that the disseminated video stream adheres to the most stringent aspect specified in the cascade of all privacy filters relevant to a video stream or query. To support this privacy framework, we extend our Live Video Database Model with an informatics-based approach to object recognition and tracking and add an intrinsic privacy model that provides a level of privacy protection not previously available for real-time streaming video data. The proposed framework also provides a formal approach to implement and enforce privacy policies that are verifiable, an important step towards privacy certification of video surveillance systems through a standardized privacy specification language.

Privacy protection vs. utility in visual data

“…These privacy-preserving techniques basically rely on image processing algorithms such as scrambling by JPEG-masking [35], in-painting [15], pixelation [22], blanking [3], replacement with silhouettes [59], blurring [36], warping or morphing [29]. In a recent workshop dedicated protection methods have been proposed in order to solve the specified visual privacy task [5,19,20,30,34,40,41,49].…”

Secure JPEG scrambling enabling privacy in photo sharing

“…A lot of research efforts in image privacy have focused on approaches to incorporate privacy protection into existing security surveillance systems and frameworks, typically, by implementing access rights management and privacy policies [2][3] [4]. Another large body of work is on the development of algorithms and methods to protect visual privacy, such as using watermarking to hide visual personal information [5], scrambling techniques to reversibly distort privacy sensitive regions [6], removal of unauthorized personnel from the video feed [7], and encoder independent geometrical-based reversible distortions [8] [9].…”

“…As long as a correct scrambling key is provided, the region can be recovered. We have implemented the proposed scrambling algorithm in both transcoder and encoder/decoder using an open source JPEG library by Independent JPEG Group 4 . The process of a multi-region scrambling and selective descrambling is illustrated in Fig.…”

Privacy-preserving photo sharing based on a public key infrastructure

“…At the technical level this work broadly breaks down into two solutions spaces: Privacy may be enforced at the source level through appropriate image or video processing and encryption techniques; [1][2][3][4][5] Or it may be enforced at the media storage and server level through appropriate enforcement of access policies. [6][7][8][9] Sharing photos through OSNs substantially extends the challenge of e↵ective privacy management and control over that encountered in "closed" surveillance and security systems: The image source (e.g., the user's device) is no longer a part of the system infrastructure; the relationship between photo owner and OSN is not subject to a system policy or third-party oversight; and access control at the OSN is at the sole discretion of the OSN.…”