Real time region of interest based chaotic image cryptosystem for IoT applications

• Published in: Multimedia tools and applications Vol. 83; no. 6; pp. 16161 - 16177
• Main Authors: Sheela, S. J., Suresh, K. V.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.02.2024; Springer Nature B.V
• Subjects: Algorithms; Cloud computing; Computer Communication Networks; Computer Science; Correlation coefficients; Data Structures and Information Theory; Encryption; Entropy (Information theory); Internet of Things; Multimedia Information Systems; Special Purpose and Application-Based Systems; Chaos; Real time image encryption; YOLOv3; MHM; ROI; Object detection
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-16093-3

Recently, internet of things (IoT) is extensively used in wide applications like industry, e-healthcare, agriculture to send images through internet for the cloud storage. However, these images should be secured before the transmission. In this context, a novel cryptosystem for region of interest protection of images is proposed in this article. The proposed cryptosystem employs You Only Look Once v3 (YOLOv3) to detect regions of interest (ROI) in the original image and a new row column shift (RCS) image encryption algorithm based on two dimensional modified Henon map (2D-MHM) in order to obtain encrypted object image. A series of analyses are carried out to validate the proposed cryptosystem which includes histogram analysis, correlation coefficient analysis, information entropy analysis etc. The experimental results reveal that the ROI based image encryption algorithm shows strong performance with respect to security. Moreover, it reduces the time and computational complexity when compared to full encryption scheme.