Paper Title
An Efficient Image Encryption Algorithm for High Security through Visual Cryptography

Abstract
Data encryption is significant during the transmission of information in networks since data loss, insider threats, phishing, and data masking are some forms of attacks that the hacker used for stealing either personal or organizational information. Particularly, images are manipulated by attackers using tools that alter the original meaning of the image. Enforcing strong standards in data security can improve resistance to access and safeguard privacy information from cyber security threats. The traditional encryption algorithms and mechanisms have resulted in security provision but have not to find compatibility in real-time implementations. Hence, the proposed study employs an Enhanced hybrid AES and ECC algorithm for the encryption and decryption process using Visual Cryptography (VC) which splits the real image into shares and encrypts for transmission. Further, the study utilizes the Deep Learning (DL) algorithms for comparing the original image with the decrypted form of the image to validate the originality of the image being transferred. The Enhanced hybridization of(CNN) Convolutional Neural Network with (LSTM) Long Short Term Memory analyses characteristics and learns from the low-level resolution of feature maps to the pixel-wise prediction for identifying the tamper in images.Hence the proposed technique identifies the image manipulations at the pixel level with better accuracy and precision which can be computed through performance analysis of the system. The performance metrics of the proposed work are being compared with the existing works like the CVC algorithm, Siamese network, and hybrid method that determines the efficiency of the proposed system. Keywords - Secure image hashing, Elliptical curve cryptography (EEC), Deep Learning (DL), Long Short Term Memory (LSTM), Advanced Encryption Standard (AES), Convolutional Neural Network (CNN), Visual Cryptography (VC)