Toward Privacy Protection of Sensed Biometric Features with Extended Visual Cryptography

The popularity of more powerful and smarter digital devices has improved the quality of life and poses new challenges to the privacy protection of personal information. Although traditional encryption algorithms are effective in most application scenarios, they may not be suitable in integrated sensing and communication environments due to their low processing capabilities. In this paper, we develop a novel extended visual cryptography scheme to store private information in separate databases, which eliminates the complex operations in privacy-preserving schemes based on cryptography and watermarking. Since shares do not reveal any feature about private sensed information, we can transmit sensitive information among sensors and smart home servers in plain. To abate the influence of noise due to the use of visual cryptography, we leverage the generalization ability of transfer learning to train a visual cryptography-based recognition network. Experimental results show that our proposed method keeps the high accuracy of the feature recognition system when providing security.