A Multiple Images Hiding And Encryption Scheme Based On Chaotic Compressive Sensing

With the rapid development of cloud computing and the Internet of Things technology,various intelligent interactive applications have increasingly penetrated into every sphere of human life,so that private data is transmitted in open data links,and stored on the shared cloud service platform.There are inevitably some potential security problems,for example,the personal private data being stolen or maliciously used,which will bring a great threat to the security of the personal information of the users and virtual property.Private data include audio,video,images,files,etc.Among them,video and images will account for a large proportion in the future with the rise of VR/AR.In recent years,there are some visual cryptography systems which are image hiding schemes essentially.The plain image is encrypted into a cipher image which is visually indistinguishable effectively,and this encryption process can reduce the risk of further attacks by the eavesdropper.However,there are limitations in all traditional encryption algorithms which cannot meet the requirements of security and computational efficiency simultaneously,and the image hiding technology is at the bottleneck of the balance between hidden capacity and imperceptibility.In view of these two problems,this paper combines compressive sensing technology and chaotic theory to design a scheme that can increase the embedded capacity and improve the security at the same time,and then we have carried out in-depth research and analysis on it.The main innovations and research results are described as follows:(1)We propose a multiple images hiding model based on compressive sensing with the large hiding capacity,and this model combines compressive sensing technology to embed the multiple images into the carrier image simultaneously,so that the embedding rate reaches 6 bits per pixel,and the proposed scheme can save more than 3/7 of the storage space.It breaks the bottleneck of the balance between hidden capacity and imperceptibility in traditional image hiding algorithms.In addition,the proposed scheme will obtain a visually meaningful cipher image if the image is reconstructed directly,which can reduce the risk of further decryption by the eavesdropper that ensure high covertness during the transmission.Through experimental simulation,we analyze the effect of the parameter conditions on the image perceptibility and the restoration,and the parameter conditions include the embedding strength,the compression ratio,and the sparse basis.Then we prove that the proposed model has higher embedding efficiency.(2)We propose a multiple images hiding encryption scheme based on chaotic compressive sensing.The scheme combines the characteristics of the chaotic system.The chaotic scrambling is applied on the discrete wavelet transforming of the plain images,and the chaotic sequence is used to generate the measurement matrix,which improves the reconstruction effect and the security of the proposed model.Numerical simulations and performance analyses present that the proposed scheme can recover the plain images better,improve the computational efficiency of multiple images' encryption and reconstruction,and it has high key sensitivity,large key space,and good robustness against noise and reduction attack.