Image processing encryption
Yükleniyor...
Tarih
2020
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
IOP Publishing Ltd
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
The aim of this research paper is to develop a new approach of image processing encryption using machine learning techniques in python So that the human cannot understand the images because it is in encrypted form and can be securely transfer to its destination. We also used the computer-generated Holography (CGH) technique to encrypt and decrypt images. We first implement an existing algorithm and verify the claims of the authors. We then investigate higher dimensional Baker maps for image encryption. For this, we first propose a new interpretation for the Baker map in terms of a path function S. We then apply the higher dimensional maps for image encryption and experimentally conclude that 3D Baker map suffices for encryption. That is, there is no perceptible performance gained when using higher dimensional Baker maps. Next, in an attempt to use chaotic maps for the diffusion mechanism in the encryption scheme, we embed the diffusion process into the confusion process. For this, we first propose an alternative view of a 2D image as a 3D structure using the binary representation of the image intensity values. We extend this scheme from grayscale images to color images and show its immense value in color image encryption. Lastly, we propose a Baker map based on random walk of the image. Here, we employ sparse decomposition of images as a method of generating the random paths. Random walk-based Baker maps would be more difficult to break than traditional Baker maps because of the chaotic behavior in the walk itself. The significance of images and their sharing is increasing day by day. Their security is becoming an important issue while transferring over a public network. To protect images from hacker's secret sharing is one of the best techniques. The secret sharing is a way to share a secret with n participants and then setup is made for t or more number of participants who must contribute to revealing the secret. Here t < n is known as a threshold which must be achieved for secret reconstruction. © Published under licence by IOP Publishing Ltd.
Açıklama
2nd International Scientific Conference of Al-Ayen University, ISCAU 2020 -- 15 July 2020 through 17 July 2020 -- -- 165187
Anahtar Kelimeler
3D Baker Map, Computer Generated Holography (CGH), Machine Learning, Secret Image Sharing
Kaynak
IOP Conference Series: Materials Science and Engineering
WoS Q Değeri
Scopus Q Değeri
N/A
Cilt
928
Sayı
2
