SMPCS: Self-Managed Privacy Control and Sharing Scheme Based on Thumbnail Preserving and ECDH for Social Networks

The rapid development of social networks, coupled with the proliferation of Internet of Things image-capture devices, has made sharing images easier but also raised concerns about privacy leakage. For this reason, image content privacy protection has attracted great attention and focused research. However, existing schemes suffer from participants inability to self-manage privacy, poor visual usability of protected images, and risks of key theft during transmission. To address these challenges, this article proposes a self-managed privacy control and sharing scheme (SMPCS) based on thumbnail-preserving and elliptic curve Diffie-Hellman (ECDH) for social networks. In SMPCS, we first design a key generation and sharing scheme supporting multikey distribution based on ECDH and public key authentication. This scheme generates a secure data block for each participant, containing their facial codes and the owner’s public key, to ensure the security of key transmission. Next, the sensitive areas in social image are encrypted using the designed adaptive thumbnail-preserving encryption based on sensitive area (ATPE-SA), ensuring that the encrypted social image maintains visual usability. Moreover, utilizing the designed adaptive blocking module, participants can balance the privacy and usability of encrypted images by adjusting privacy parameters to meet diverse wishes. Finally, participants can choose whether to reconstruct social images according to their privacy wishes, achieving autonomous privacy control. Extensive experimental analyses and comparisons with state-of-the-art schemes demonstrate the superiority of the proposed SMPCS in terms of autonomous privacy control, usability and security.