支持密文搜索和运算的遥感图像加密研究
发布时间：2018-07-24 14:14
【摘要】：遥感图像作为一种重要的时空数据,具有多精度、多时相、多语义、多波段等特点,在土地规划、环境监测、灾害预报等重大领域已发挥出了令人瞩目的潜能及优势,尤其在信息获取及分析处理方面展示了其独特的魅力。随着海洋立体观测技术的飞速发展,催生了高精度、大范围的遥感图像数据呈爆炸式增长,通过不同途径获取到的遥感图像数据每天以TB级的速度迅猛增长,目前已达到EB级别,已展示出了明显的“大数据”特征,为各行各业的科研和发展提供强大的数据支撑。目前的单机难以处理具有大数据特征的遥感图像,因此需要利用云环境的优势进行高效率的计算。作为重要的基础决策性信息资源,遥感图像在军事行业具有举足轻重的决策地位,目前已成为最有效、最安全,同时也是最可靠的侦察手段,可对作战区域进行实时的全方位遥感侦察。云非完全可信,安全事件频发,鉴于遥感图像在民生行业及国防军事中的重要性,为防止云环境下受到恶意的截取或破坏,亟需可靠的技术手段对其进行加密保护,使得在云环境下不仅能够保证密文遥感图像的传输、存储以及共享,而且能够支持密文的计算,提高云环境下密文遥感图像的有效性。云环境下海量密文遥感图像的搜索和运算,其运算效率及搜索精度对遥感图像的广泛使用具有直接性的意义。由于目前大多数可计算加密算法并不支持在密文遥感图像上进行搜索与运算,因此本文针对遥感图像的属性特征,提出了支持密文搜索和运算的遥感图像加密方案,使得加密后的遥感图像能够支持密文的搜索与运算。本文提出的可计算加密算法主要体现以下两个方面:1、本文提出了一种支持密文搜索的遥感图像加密方案,根据遥感图像的成像原理及多波段、高精度等特性,利用基于排序的二维非线性映射对遥感图像进行空域加密,其次采用改进的Henon映射对遥感图像各波段进行灰度加密。根据遥感图像多波段特性,通过统计灰度值区间信息构造密文遥感图像的特征向量,最后根据相似度匹配算法搜索目标图像。实验采用Landsat 8遥感图像并对其进行加密分析,实验表明该加密方案在非完全可信云环境下保证遥感图像安全的同时,有效地提高了密文遥感图像搜索的效率,本地用户只需要一次加解密时间,通信成本开销小,且算法计算复杂度低。2、本文提出的基于最重要位平面(MSB,Most Significant Bit)的同态加密方案,在云环境下不仅能够实现密文遥感图像的安全保护,同时能够对加密的遥感图像进行有效地线性操作。结合遥感图像的MSB信息,利用基于整数的矩阵同态加密算法,使得加密后遥感图像的灰度值均匀分布在高频区域,很好的隐藏了明文图像的特征信息,在非完全可信云环境下保证了密文数据的安全。加密后的遥感图像能够直接支持线性操作,通过波段间线性插值运算能得到同一区域不同时间点遥感图像的差异度效果图,此外还可以结合边缘检测算子在密文上直接获得加密的图像边缘信息,很好的提高了云环境下密文遥感图像的可用性。综上所述,本文对非可信云环境下支持密文可计算的加密算法展开了系统研究,提出的加密方案能够很好的支持密文遥感图像的搜索及运算操作,实验证明该加密算法在遥感领域具有重要的理论研究意义和工程应用价值。
[Abstract]:As an important spatio-temporal data, remote sensing image has the characteristics of multi precision, multi time phase, multi semantics, multi band and so on. It has shown great potential and advantages in land planning, environmental monitoring, disaster prediction and other major fields, especially in information acquisition and analysis. The rapid development of technology has brought about high precision. The remote sensing image data of large range has been explosively growing. The remote sensing image data obtained through different ways are growing rapidly at the speed of TB level every day. Now it has reached the level of EB, which has shown obvious "big data" features and provides powerful data for scientific research and development of all walks of life. It is difficult to handle remote sensing images with large data features, so it is necessary to make use of the advantages of the cloud environment for efficient calculation. As an important basic decision-making information resource, remote sensing images have a decisive position in the military industry, and are now the most effective, safest, and the most reliable detection. In view of the importance of remote sensing images in the people's livelihood industry and national defense and military, a reliable technical hand is urgently needed to encrypt and protect it in the cloud environment, which makes it not only in the cloud environment, but also in the cloud environment. It can guarantee the transmission, storage and sharing of the ciphertext remote sensing images, and can support the calculation of the ciphertext and improve the effectiveness of the remote sensing images in the cloud environment. The search and operation of the massive encrypted remote sensing images under the cloud environment are of direct significance to the extensive use of remote sensing images. Most computable encryption algorithms do not support the search and operation of the ciphertext remote sensing images. Therefore, in this paper, a remote sensing image encryption scheme is proposed to support the search and operation of ciphertext for the properties of remote sensing images. The encrypted remote sensing image can support the search and operation of the ciphertext. The computable encryption algorithm proposed in this paper is proposed. It mainly embodies the following two aspects: 1, this paper proposes a remote sensing image encryption scheme that supports ciphertext search. According to the imaging principle of remote sensing image and the characteristics of multi band and high precision, the remote sensing image is encrypted by the sort based two-dimensional nonlinear mapping. Secondly, the improved Henon mapping is used for the remote sensing images. According to the multi band characteristics of remote sensing image, the feature vectors of the encrypted remote sensing image are constructed by the statistical gray value interval information. Finally, the target image is searched according to the similarity matching algorithm. The experiment uses Landsat 8 remote sensing image and encrypts it. The experimental table shows that the encryption scheme is guaranteed under the non fully trusted cloud environment. At the same time, it can effectively improve the efficiency of remote sensing image search. Local users only need one time of encryption and decryption, the cost of communication is small, and the computational complexity of the algorithm is low.2. The homomorphic encryption scheme based on the most important bit plane (MSB, Most Significant Bit) proposed in this paper can not only be implemented in the cloud environment. The security of the encrypted remote sensing image can be protected effectively. At the same time, the encrypted remote sensing image can operate effectively and linearly. Combining the MSB information of remote sensing images, using the integer based matrix homomorphic encryption algorithm, the gray value of the remote sensing image is evenly distributed in the high frequency region after the encrypted image, and the feature information of the plaintext image is hidden well, and it is not finished. The encrypted remote sensing image can directly support linear operation and can get the difference effect map of the remote sensing images of the same region at different time points through the linear interpolation operation between the bands. In addition, the encrypted image edge message can be directly obtained on the ciphertext by the edge detection operator. In summary, this paper studies the encryption algorithm that supports cipher computable encryption in the untrusted cloud environment. The proposed encryption scheme can support the search and operation of the encrypted image very well. The experiment proves that the encryption algorithm has the advantages of the remote sensing field. Important theoretical research significance and engineering application value.
【学位授予单位】：上海海洋大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP751
[Abstract]:As an important spatio-temporal data, remote sensing image has the characteristics of multi precision, multi time phase, multi semantics, multi band and so on. It has shown great potential and advantages in land planning, environmental monitoring, disaster prediction and other major fields, especially in information acquisition and analysis. The rapid development of technology has brought about high precision. The remote sensing image data of large range has been explosively growing. The remote sensing image data obtained through different ways are growing rapidly at the speed of TB level every day. Now it has reached the level of EB, which has shown obvious "big data" features and provides powerful data for scientific research and development of all walks of life. It is difficult to handle remote sensing images with large data features, so it is necessary to make use of the advantages of the cloud environment for efficient calculation. As an important basic decision-making information resource, remote sensing images have a decisive position in the military industry, and are now the most effective, safest, and the most reliable detection. In view of the importance of remote sensing images in the people's livelihood industry and national defense and military, a reliable technical hand is urgently needed to encrypt and protect it in the cloud environment, which makes it not only in the cloud environment, but also in the cloud environment. It can guarantee the transmission, storage and sharing of the ciphertext remote sensing images, and can support the calculation of the ciphertext and improve the effectiveness of the remote sensing images in the cloud environment. The search and operation of the massive encrypted remote sensing images under the cloud environment are of direct significance to the extensive use of remote sensing images. Most computable encryption algorithms do not support the search and operation of the ciphertext remote sensing images. Therefore, in this paper, a remote sensing image encryption scheme is proposed to support the search and operation of ciphertext for the properties of remote sensing images. The encrypted remote sensing image can support the search and operation of the ciphertext. The computable encryption algorithm proposed in this paper is proposed. It mainly embodies the following two aspects: 1, this paper proposes a remote sensing image encryption scheme that supports ciphertext search. According to the imaging principle of remote sensing image and the characteristics of multi band and high precision, the remote sensing image is encrypted by the sort based two-dimensional nonlinear mapping. Secondly, the improved Henon mapping is used for the remote sensing images. According to the multi band characteristics of remote sensing image, the feature vectors of the encrypted remote sensing image are constructed by the statistical gray value interval information. Finally, the target image is searched according to the similarity matching algorithm. The experiment uses Landsat 8 remote sensing image and encrypts it. The experimental table shows that the encryption scheme is guaranteed under the non fully trusted cloud environment. At the same time, it can effectively improve the efficiency of remote sensing image search. Local users only need one time of encryption and decryption, the cost of communication is small, and the computational complexity of the algorithm is low.2. The homomorphic encryption scheme based on the most important bit plane (MSB, Most Significant Bit) proposed in this paper can not only be implemented in the cloud environment. The security of the encrypted remote sensing image can be protected effectively. At the same time, the encrypted remote sensing image can operate effectively and linearly. Combining the MSB information of remote sensing images, using the integer based matrix homomorphic encryption algorithm, the gray value of the remote sensing image is evenly distributed in the high frequency region after the encrypted image, and the feature information of the plaintext image is hidden well, and it is not finished. The encrypted remote sensing image can directly support linear operation and can get the difference effect map of the remote sensing images of the same region at different time points through the linear interpolation operation between the bands. In addition, the encrypted image edge message can be directly obtained on the ciphertext by the edge detection operator. In summary, this paper studies the encryption algorithm that supports cipher computable encryption in the untrusted cloud environment. The proposed encryption scheme can support the search and operation of the encrypted image very well. The experiment proves that the encryption algorithm has the advantages of the remote sensing field. Important theoretical research significance and engineering application value.
【学位授予单位】：上海海洋大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP751
【参考文献】
相关期刊论文 前10条
1 杨艺芳;王宇平;;基于核模糊相似度度量的谱聚类算法[J];仪器仪表学报;2015年07期
2 李顺东;窦家维;王道顺;;同态加密算法及其在云安全中的应用[J];计算机研究与发展;2015年06期
3 苏sコ，
本文编号：2141633
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2141633.html
