Font Size: a A A

Remote Sensing Image Encryption Supporting Ciphertext Search And Computation

Posted on:2018-08-30Degree:MasterType:Thesis
Country:ChinaCandidate:X GengFull Text:PDF
GTID:2348330536977347Subject:Computer technology
Abstract/Summary:PDF Full Text Request
As an important spacetime data,remote sensing image has the character of multi-precision,multi-temporal,multi-temporal,multi-semantic,multi-band and so on.And it plays a remarkable potential and advantages in the land area planning,environmental monitoring,disaster prediction and other major fields,especially in information access and analysis,it has shown different charming.With the rapid development of comprehensive observation technology,the remote sensing data has exploded quickly with the characteristics of high-accuracy,high-frequency and high-range,and it grows rapidly through different routes in TB level for every day,and nowadays it has been recognized as “big data” and acted as data supporting in all kinds of industries.At present,it is difficult to deal with remote sensing image in stand-alone mode,so it is necessary to use the advantage of cloud environment to calculate efficiently.As an important strategic and basic information resource,remote sensing image plays an important role in making decision in the military field,nowadays it has become the most effective secure measures,and it is the most reliable means of reconnaissance.The remote sensing image can reconnoiter the combat area for all days,and it keeps high dynamic aerospace for all domains.The cloud is non-fully trust and the security incidents break out frequently.In view of the importance of remote sensing image in the people's livelihood industry and defense military,there is a need for reliable technical means to encrypt the remote sensing image from being intercepted or destroyed in the process of transmission,storage and sharing in the non-fully trusted cloud environment,and the encryption algorithm not only guarantee the security of the ciphertext remote sensing image,but also can support the computable operation of the ciphertext remote sensing image,in some times the algorithm improves the effectiveness of the ciphertext remote sensing image in cloud.The efficiency and precision of the remote sensing image directly affect the widely use in search and computation.Recently,most of the computable encryption algorithms can't be directly supported for search and computation on encrypted remote sensing image,so the paper proposes an encryption algorithm which is based on the character of remote sensing image,and the encrypted remote sensing image can support the search and linear operation.In the paper,the computable encryption algorithm mainly embodies the following two aspects:1.The paper proposes a searchable encryption scheme for cipher remote sensing image.According to the imaging principle and multi-band feature of remote sensing image,the scheme encrypts the airspace of remote sensing image with the two-dimensional non-linear mapping,and then it encrypts the gray value of each band with the advanced Henon mapping.The feature vector of the remote sensing image is constructed by the interval information of the gray value,and the target image is searched according to the similarity matching algorithm.The experiment takes the remote sensing image of Landsat 8,and then take the ciphertext into analysis,the experiment shows that the encrypted scheme not only can support the security of remote sensing image in non-fully trusted cloud environment,but also can improve the efficiency of ciphertext search,the user only need one time to encrypt and decrypt the remote sensing image,and it causes the low cost and computational complexity.2.Based on most significant bit of remote sensing image,the paper proposes a homomorphic encryption scheme,which not only can realize the security protection of ciphertext remote sensing image in cloud,but also can support the effective linear operation in the encrypted remote sensing image.Combining the most significant bit of the remote sensing image and taking advantage of matrix homomorphic encryption algorithm,the encrypted gray value of ciphertext is distributed in the high frequency,and it is difficult to obtain the sensitive features of the plaintext through the gray information.It is finely improved the security and usability of the ciphertext remote sensing image in the non-fully trusted cloud environment.The ciphertext of remote sensing image can support the liner operation,through the experimental simulation of remote sensing image with different time in a fixed strait region,the encryption algorithm can conceal the character information of the plaintext well.In addition,it is easy to obtain the edge information of remote sensing image directly on the ciphertext,and it is finely to improve the usability of the remote sensing image.In summary,the paper has taken the cryptographic algorithm into research which can support ciphertext computing in non-fully trusted cloud environment.The proposed encryption scheme can finely support the search and linear operation in ciphertext remote sensing image.Experiments show that the encryption algorithm has important theoretical significance and practical engineering applications value in the field of remote sensing.
Keywords/Search Tags:computable encryption, remote sensing image, Henon mapping, feature vector, cloud computing, homomorphic encryption
PDF Full Text Request
Related items