Minor Privacy Protection Through Real-time Video Processing at the Edge

Abstract

The collection of a lot of personal information about individuals, including the minor members of a family, by closed-circuit television (CCTV) cameras creates a lot of privacy concerns. Particularly, revealing children's identifications or activities may compromise their well-being. In this paper, we investigate lightweight solutions that are affordable to edge surveillance systems, which is made feasible and accurate to identify minors such that appropriate privacy-preserving measures can be applied accordingly. State of the art deep learning architectures are modified and re-purposed in a cascaded fashion to maximize the accuracy of our model. A pipeline extracts faces from the input frames and classifies each one to be of an adult or a child. Over 20,000 labeled sample points are used for classification. We explore the timing and resources needed for such a model to be used in the Edge-Fog architecture at the edge of the network, where we can achieve near real-time performance on the CPU. Quantitative experimental results show the superiority of our proposed model with an accuracy of 92.1% in classification compared to some other face recognition based child detection approaches.

Privacy-Preserving Surveillance as an Edge Service Based on Lightweight Video Protection Schemes Using Face De-Identification and Window Masking

“…At present, CCTV cameras are abundantly deployed in many parts of urban areas by governments, companies, and individuals with the main goal of ensuring physical security and safety. For example, an average Londoner is estimated to be caught on CCTV cameras 300 times a day while moving around the city [5,9,10,13,14]. As a result, a great deal of information about individuals is collected by the CCTV cameras without their knowledge and consent.…”

“…The UACI is defined by Equation ( 13), which is employed to measure the average intensity difference in a color channel between its two cipher versions C 1 (i, j) and C 2 (i, j). Besides, the mathematical definition of the NPCR, measures the change rate of the number of pixels of the cipher-frame when only a bit of the original key or pixel is modified, is provided by Equation (14).…”

Secure and Privacy-Preserving Stored Surveillance Video Sharing atop Permissioned Blockchain

“…Today, more than a billion of CCTV cameras are pervasively deployed around the world in urban areas by governments, companies, and individuals with the main goals of ensuring physical security and public safety. In places like Beijing, London, and New York, an individual is estimated to be caught on CCTV cameras hundreds of times a day [9], [12], [26], [30]. Consequently, a tremendous amount of information about individuals is garnered without their knowledge and consent.…”