Adaptive Secure Image Transmission Using Tri-Layered Visual Cryptography and Residual Error Reduction for AI-Driven Interactive Mobile Learning Platforms.

Saved in:

Bibliographic Details
Title:	Adaptive Secure Image Transmission Using Tri-Layered Visual Cryptography and Residual Error Reduction for AI-Driven Interactive Mobile Learning Platforms.
Authors:	Raju, S. Shibani¹ shibanir@srmist.edu.in, Thenmozhi, R.¹ thenmozr@srmist.edu.in
Source:	International Journal of Interactive Mobile Technologies. 2026, Vol. 20 Issue 12, p72-86. 15p.
Subjects:	Visual cryptography, Image reconstruction, Image quality analysis, Information technology security, Deep learning, Mobile learning
Abstract:	As interactive mobile learning platforms based on artificial intelligence (AI) are being adopted rapidly, secure and efficient visual educational data communication have emerged as a crucial challenge. In this case, it is still important not to disclose any sensitive learning material (e.g., assessments, medical images and/or proprietary instructional media) while ensuring high visual fidelity for effective learning throughout the design process. In this study, we put forward an Adaptive Secure Image Transmission model using a tri-layered visual cryptography (TLVC) with the assistance of residual error reduction (RER) process. As such, the proposed system adopts a three-layer visual cryptographic method that breaks up input images into several insecure shares in which no single share contains anything of value by itself. Mobile-specific: Share generation can be dynamically tuned to the capability of device CPU, memory, network factors such as latency or available bandwidth. In order to combat more quality degradation caused by reconstruction, a RER module combines with lightweight deep residual learning is incorporated to improve the clarity of images and restore fine-grained structures. In addition, the framework's integration with AI-based learning platforms allows for fast content delivery, interactive visualization of data, and privacy-preserving analytics that can be aggregated across multiple sources. Experimental evaluations prove that the proposed method guarantees a better security robustness and a more accurate reconstructed image in terms of mean squared error and peak signal-to-noise ratio in comparison to classical visual cryptography. The system also has low computational overhead and is adaptable to resource-scarce mobile devices. [ABSTRACT FROM AUTHOR]
	Copyright of International Journal of Interactive Mobile Technologies is the property of International Journal of Interactive Mobile Technologies and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	As interactive mobile learning platforms based on artificial intelligence (AI) are being adopted rapidly, secure and efficient visual educational data communication have emerged as a crucial challenge. In this case, it is still important not to disclose any sensitive learning material (e.g., assessments, medical images and/or proprietary instructional media) while ensuring high visual fidelity for effective learning throughout the design process. In this study, we put forward an Adaptive Secure Image Transmission model using a tri-layered visual cryptography (TLVC) with the assistance of residual error reduction (RER) process. As such, the proposed system adopts a three-layer visual cryptographic method that breaks up input images into several insecure shares in which no single share contains anything of value by itself. Mobile-specific: Share generation can be dynamically tuned to the capability of device CPU, memory, network factors such as latency or available bandwidth. In order to combat more quality degradation caused by reconstruction, a RER module combines with lightweight deep residual learning is incorporated to improve the clarity of images and restore fine-grained structures. In addition, the framework's integration with AI-based learning platforms allows for fast content delivery, interactive visualization of data, and privacy-preserving analytics that can be aggregated across multiple sources. Experimental evaluations prove that the proposed method guarantees a better security robustness and a more accurate reconstructed image in terms of mean squared error and peak signal-to-noise ratio in comparison to classical visual cryptography. The system also has low computational overhead and is adaptable to resource-scarce mobile devices. [ABSTRACT FROM AUTHOR]
ISSN:	18657923
DOI:	10.3991/ijim.v20i12.62162