SIFT-based blockchain model for preventing the spread of forgery image
Hao Liu, Xueqing Zhao, Xin Shi, Yun Wang, Guigang Zhang, Yibing Chen · 2023
The detection of forgery images has become a hot issue in the field of communication. It is of great importance on researching how to prevent the uploading of forged images from the source of propagation. Blockchain can effectively protect the original features of images through distributed ledgers, consensus algorithms, asymmetric encryption, and smart contracts. Therefore, this paper proposes an idea of preventing the forgery images from the very beginning on the basis of blockchain, which aims at storing the original images into the blockchain through interaction with smart contracts. Our method not only provides safe and traceable copyright protection to the owners of the original images but also effectively suppresses the uploading of forgery images. Firstly, image features are extracted by Scale-invariant feature transform algorithm(SIFT) for image to be uploaded, the features are further matched by Fast Library for Approximate Nearest Neighbors(FLANN) with with the stored images in the blockchain. Secondly, the similarity between images is calculated in order to determine the originality of uploaded images, the ones with high similarity scores are refused to upload into the blockchain. At the same time, the copyright owner of the infringed image will be informed immediately. Finally, a large number of simulation results show that the method proposed in this paper can effectively authenticate the image copyright and prevent the spread of forgery images.