A secure image evidence management framework using multi-bits watermark and blockchain in IoT environments

Criminal forensics in an Internet-of-Things (IoT) environment often requires complex investigations because IoT devices usually generate a large amount of electronic data, especially image data, which brings great difficulties to traditional digital forensic methods. Therefore, designing a secure management solution for image data and making it to be image evidence available in court is a new challenge. In this paper, we present a new secure image evidence management framework based on multi-bits digital watermarking and blockchain. In this framework, we first propose a flexible and robust self-learning based watermark embedding algorithm, which can embed both image marks and binary messages into the latent space of the input images and also improve the resistance of a broad range of attacks (geometric transform, JPEG, noise, etc.) to the watermarked images. And then, we design a smart contract resided in blockchain, which can achieve safe storage and automatically authentication of embedded watermarks. The experimental results on both watermarking algorithm and smart contracts have demonstrated the feasibility and efficacy of the proposed framework.