Research On Image Storage Authentication System Based On Blockchain

Image copyright protection is usually achieved by digital watermarking,which is also widely used in the field of data protection and data ownership validation.However,existing digital watermarks have the inherent disadvantage of requiring a fully trusted third party to act as arbiter,which can be difficult to achieve.Secondly,image watermarking algorithms inevitably result in the loss of image data when they operate on it,which is usually irreversible.Zero-watermarking algorithms can solve the problem of data loss,but they are more dependent on trusted third parties than traditional digital watermarking,which makes their prospects limited.Overall,the main tasks of this paper are:(1)Based on the Ethereum smart contract technology,the corresponding image deposition process is designed,and a decentralised image deposition and validation mechanism is constructed for practical application.The prototype system is built on this mechanism,and the design of a smart contract for the authentication of image ownership and the transaction of usage rights is completed,based on the automatic execution and immutability of the smart contract,the change of image ownership and the transaction of usage rights are realised,and the transaction history is recorded with a certain status value,and the contract has the function of storing and querying the image ownership.This study solves the problem of traditional digital watermarking relying on trusted third parties and protects the original image as well by introducing IPFS as part of the overall solution,making the whole process simpler and faster.A highly robust zero watermarking algorithm was designed to address the large storage space,diversity and sparse distribution of individual images of image data,while causing as little data loss as possible to the original image.An image normalisation technique is used to convert the image to a standard image in order to enhance the robustness of the algorithm against geometric attacks.The low frequency coefficients obtained from the quadratic wavelet transform concentrate most of the energy of the image,while the maximum singular value obtained from the block singular value decomposition has good stability,further enhancing the robustness of the algorithm.(2)Current consensus algorithm research focuses on improving information secure interaction and greater throughput,but there are few consensus algorithms designed for image protection,so this study improves the PBFT algorithm for the federated chain and proposes the Practical Byzantine Fault Tolerance for Image protection(IPBFT)algorithm,enabling it to improve the efficiency and reduce the cost of storage while maintaining security.Building a bridge between PBFT and image protection ensures effective image copyright protection and secure transactions.Use a node lightweighting(offloading solution)for cases where the image files and watermark files are large to affect the space utilization and time complexity of the blockchain,consider offloading redundant images to IPFS without affecting the normal validation of the blocks and the normal operation of the blockchain.Improvements are proposed for fast transactions,with the normal consensus algorithm taking the approach of verifying before uploading to the chain,while the IPBFT algorithm uploads before verifying.To ensure security,a set of node credit value scoring criteria is also designed.Nodes that choose fast transactions have to pay a deposit based on their credit,and too low a credit will not be able to choose swift transactions.