Research On Asymmetric Digital Fingerprint Technology Of Remote Sensing Images

With the rapid development of remote sensing image application technology,remote sensing images are playing an increasingly important role in many fields such as national defense construction,military construction,and modernization construction,and their social and economic value has become increasingly prominent.However,with the widespread application of remote sensing images,the problem of image copyright protection has become more and more serious.As an important carrier of spatial feature information in the geographic information field,solving the problem of digital copyright protection of remote sensing images has caused a wide range of scholars and content providers.attention.As an emerging digital copyright protection technology,digital fingerprints can provide new ideas for image copyright protection.Compared with traditional digital watermarking,digital fingerprints have the function of "tracing illegal disseminators".This research will focus on the targeted application of digital fingerprint technology in remote sensing images,and strive to provide new feasible solutions for the copyright protection of remote sensing images.The basic idea of digital fingerprinting is to embed a unique fingerprint sequence(array,code,string,etc.)into the data during the distribution of digital files(or copy).This fingerprint not only has the ability to identify copyright,but also It also has the ability to track pirates.If the content provider of the image finds that there is an unauthorized digital image file in the Internet environment,it can effectively track the pirate through the two steps of fingerprint extraction and fingerprint identification.In order to ensure the validity and non-repudiation of the image copyright authentication process,and prevent the occurrence of framing,it is an effective way to design an "asymmetric digital fingerprint" suitable for remote sensing images.Asymmetric digital fingerprint algorithms have been widely used in multimedia fields(such as digital images,video,audio,etc.).However,no digital fingerprint copyright protection algorithm for remote sensing images has been proposed.Aiming at the sufficient advantages of the aforementioned asymmetric digital fingerprints in content copyright protection and the lack of relevant targeted application algorithms,this paper proposes two asymmetric digital fingerprint schemes for remote sensing images.The specific work is as follows:(1)The applicability of asymmetric digital fingerprints on remote sensing images is analyzed.By comparing the existing mainstream asymmetric digital fingerprint schemes and comprehensively analyzing the characteristics of remote sensing images,an asymmetric digital fingerprint scheme suitable for remote sensing images is obtained.Based on this,the following two specific types are used for copyright protection and collusion.The asymmetric digital fingerprint scheme of remote sensing images for user tracking lays the foundation.(2)Aiming at the problems of high computational complexity of asymmetric digital fingerprints and low network bandwidth efficiency,an asymmetric digital fingerprint scheme suitable for remote sensing images based on homomorphic public key encryption is proposed.In this scheme,the content provider encrypts remote sensing images through the DCT spread spectrum scrambling method,and uses Bresson homomorphic public key encryption to achieve asymmetric distribution of decryption keys.At the client,the remote sensing image containing the digital fingerprint is decrypted.In the decryption stage,using different decryption keys can generate different fingerprint-containing remote sensing image copies.The public key encryption algorithm in the scheme does not directly encrypt remote sensing images,thereby reducing computational complexity and improving encryption efficiency;at the same time,because the content provider only needs to generate the same fingerprint copy for multiple consumers,and transmit it through multicast transmission.Distributed to different consumers,so the bandwidth requirements are reduced.Experiments show that in the case of a large number of users,this scheme can effectively improve bandwidth efficiency and encryption efficiency,significantly reduce the computing load of the data server,and reduce user waiting time.(3)In order to efficiently track the co-conspirators in the distribution of remote sensing images,a remote sensing method based on Optimal Focused Fingerprints from Orthogonality(OFFO)and Balanced Incomplete Block Design(BIBD)is proposed.Image anti-collusion asymmetric fingerprint scheme.This scheme takes OFFO fingerprint as the base vector prototype,quantizes it into a binary fingerprint,and then uses BIBD to encode a new multilayer fingerprint scheme.At the same time,the Bresson homomorphic public key encryption algorithm is used to realize the asymmetric distribution of the scheme.The client decrypts the remote sensing image copy with fingerprints,and different fingerprint copies come from different decryption keys.Experiments show that under the same user capacity,the scheme has better resistance to collusion attacks,and at the same time has lower bandwidth requirements and higher encryption efficiency.