In this paper, we propose a digital image authentication model with an alliance chain through blockchain + InterPlanetary File System (IPFS) technology and ResNet-50 convolutional neural network digital watermarking technology. The blockchain + distributed storage method is used to solve the problem of large-scale data uplink, and the original data is stored in the IPFS distributed system. After the image is uploaded, the image digital works are embedded through the deeplearning model based on ResNet-50 to form a new carrier digital image embedded with a watermark. When infringement is found, watermark extraction is performed on the infringing image to obtain the original watermark. By analyzing the information recorded in the blockchain and tracing the infringement on the blockchain, the copyright of users’ digital images can be protected.
Xu Y, Qian Q, 2022, Blockchain-based Data Copyright Protection and Combinatorial Bidding. Journal of Shanghai University (Natural Science Edition), 28(3): 413–426.
Li L, 2018, Research and Application of Blockchain Technology in Digital Copyright, dissertation, Northern University of Technology.
Liu Y, Zhang J, Wu S, et al., 2021, Research on Digital Copyright Protection Based on the Hyperledger Fabric Blockchain Network Technology. PeerJ Comput Sci, 7: e709. https://www.doi.org/10.7717/peerj-cs.709
Xu ZJ, Wang ZZ, Lu Q, 2011, Research on Image Watermarking Algorithm Based on DCT. Procedia Environmental Sciences, 10: 1129–1135. Proceedings of the 2011 3rd International Conference on Environmental Science and Information Application Technology ESIAT 2011.
Pradhan C, 2020, Robust and Blind Watermarking Using Arnold 4D Cat Map in Discrete Wavelet. International Journal of Information Technology, 12(2): 593–597.
Yorozu T, Hirano M, Oka K, et al., 1987, Electron Spectroscopy Studies on Magneto-Optical Media and Plastic Substrate Interface, Electron Spectroscopy Studies on Magneto-Optical Media and Plastic Substrate Interface, 2(8): 740–741.
Young M, 1989, The Technical Writer’s Handbook. Mill Valley, University Science, California.