Open Access

ARTICLE

Improving the Transmission Security of Vein Images Using a Bezier Curve and Long Short-Term Memory

Ahmed H. Alhadethi^1,*, Ikram Smaoui², Ahmed Fakhfakh³, Saad M. Darwish⁴

1 Computer Science and School of Electronics and Telecommunications, University of Sfax, Sfax, 3029, Tunisia
2 LETI Laboratory, University of Sfax, Sfax, 3029, Tunisia
3 Digital Research Center of Sfax, Technopark of Sfax, Sfax, 3029, Tunisia
4 Department of Information Technology, Institute of Graduate Studies and Research, Alexandria University, Alexandria, Egypt

* Corresponding Author: Ahmed H. Alhadethi. Email:

(This article belongs to the Special Issue: Advances and Applications in Signal, Image and Video Processing)

Computers, Materials & Continua 2024, 79(3), 4825-4844. https://doi.org/10.32604/cmc.2024.047852

Received 20 November 2023; Accepted 22 April 2024; Issue published 20 June 2024

Abstract

The act of transmitting photos via the Internet has become a routine and significant activity. Enhancing the security measures to safeguard these images from counterfeiting and modifications is a critical domain that can still be further enhanced. This study presents a system that employs a range of approaches and algorithms to ensure the security of transmitted venous images. The main goal of this work is to create a very effective system for compressing individual biometrics in order to improve the overall accuracy and security of digital photographs by means of image compression. This paper introduces a content-based image authentication mechanism that is suitable for usage across an untrusted network and resistant to data loss during transmission. By employing scale attributes and a key-dependent parametric Long Short-Term Memory (LSTM), it is feasible to improve the resilience of digital signatures against image deterioration and strengthen their security against malicious actions. Furthermore, the successful implementation of transmitting biometric data in a compressed format over a wireless network has been accomplished. For applications involving the transmission and sharing of images across a network. The suggested technique utilizes the scalability of a structural digital signature to attain a satisfactory equilibrium between security and picture transfer. An effective adaptive compression strategy was created to lengthen the overall lifetime of the network by sharing the processing of responsibilities. This scheme ensures a large reduction in computational and energy requirements while minimizing image quality loss. This approach employs multi-scale characteristics to improve the resistance of signatures against image deterioration. The proposed system attained a Gaussian noise value of 98% and a rotation accuracy surpassing 99%.

---

Keywords

---