Researches On Secure Outsourcing Searching Schemes Based On Multi-Size Encrypted Remote Sensing Images

With the rapid development of remote sensing technology,remote sensing images can provide abundant information of ground objects.At present,remote sensing image has become an important data source for dynamic disaster monitoring,ground objects change recording and real-time positioning.It plays an indispensable role in many fields,such as disaster assessment and disaster prediction,weather forecast,air quality monitoring,ecological environment monitoring,military and national defense,and becomes an important way of dynamic data observation in the global scope.However,in the field of image searching,how to search the target image securely and efficiently is a topic of concern.Remote sensing images have the characteristics of 16-bit deep gray value and many bands.As a result,using the traditional single-machine mode to process remote sensing images will cause various problems such as insufficient storage space,long processing time and poor concurrency,which cannot meet the needs of researchers for processing massive remote sensing images.In general,image searching technology requires hardware devices with high computing power to process feature vectors,Euclidean distances and other computationally intensive operations and a large number of images local storage will cause huge pressure on the cloud server.Aim to solve the two problems of image computation and image storage,it turns to the cloud computing,which is the trend of technology development.However,remote sensing images include an important strategic resource related to national security and national economic construction,satellite remote sensing images generally involve sensitive areas of a country such as oceans and forests,which are important confidential information of the state.If the remote sensing images are directly outsourced to a non-fully trusted cloud platform for image searching,it may inevitably reveal private information,and lead to illegaleavesdropping attacks during the transmission.To ensure the security of remote sensing image,it is necessary to encrypt the remote sensing images before stored on cloud platform.However,the traditional encryption methods cannot further process after encryption.Therefore,it is urgent to design an outsourcing searching solution to allow the cloud platform to search on encrypted remote sensing images.In view of the above analysis,this paper studies the secure outsourcing searching schemes from the encrypted remote sensing images,which realizes the encryption of remote sensing images and then transmits them to the cloud platform for image searching.The main innovations are as follows:1)To prevent the leakage of sensitive information in remote sensing images,a secure outsourcing searching scheme for the same size remote sensing images was proposed.Firstly,the remote sensing image is split into several bands,and the single-band image is randomly replaced by blocks and randomly replaced by pixels.The image of randomly replaced by blocks is JL converted and uploaded to the remote sensing image in cloud platform for matching cloud database.The image of randomly replaced by pixels is XOR operated,which is convenient for reversible operation to decrypt remote sensing image.To improve the security of the scheme and eliminate the strong correlation between the pixel points of the image and the surrounding points,the image pixels are rearranged by the scan mode encryption before the remote sensing image is encrypted and JL converted to disturb the image pixel position.To improve the searching efficiency,this paper randomly selects feature points,calculates their nearest distance and sub-nearest distance respectively,and regards the ratio of the two distances as a standard index to judge whether can match the feature points.according to the index,many mismatched results are removed to improve the searching efficiency of the image,and the average value of multiple point ratios is extracted to improve the accuracy.The experimental results show that the scheme has the advantages in fast searching speed,high searching efficiency and low computational complexity which improves the availability of remote sensing image secure outsourcing.2)An secure outsourcing searching scheme for multi-size remote sensing images is proposed.In view of the high requirement of the secure outsourcing searching scheme for the same size remote sensing image,it can not be applied to remote sensing images with different sizes and rotations.This paper proposes a secure outsourcing searching scheme for multi-size remote sensing images.Improved Paillierencryption combined with quantized SIFT method is used to encrypt the images,and a new two-dimensional mapping method is used to search the encrypted images.The experimental results show that the security problem of remote sensing image encryption storage and searching is well solved,and it saves the searching time and improves the efficiency while ensuring the security of remote sensing images.In this paper,we select seven bands of Landsat 8 remote sensing images with the resolution of 30 m as experimental images.The experimental results show that the secure outsourcing searching scheme of the same size remote sensing image can better realize the searching on the encrypted remote sensing image.The security and searching efficiency of the scheme are significantly improved,the searching efficiency is 98.37% which is higher than homomorphic encryption.The advantages of the scheme are high efficiency and low computational complexity,which can improve the availability of encrypted remote sensing image.The image secure outsourcing searching scheme not only effectively guarantees the image security,but also realizes the encryption searching technology of remote sensing images with different sizes.Under the condition of ensuring the security of remote sensing images,the searching time can be saved up to 96.6% compared with the original scheme and significantly improve the searching efficiency.