Hiding algorithm based fused images and Caesar cipher with intelligent security enhancement

ABSTRACT


INTRODUCTION
Due to the open nature of the internet, users face the danger of having their private information intercepted or distorted while taking advantage of the convenient electronic information exchange made possible by highly evolved internet and information technology.Thus, it is crucial to create secure communication methods [1].The most important methods for information security are ciphering and information hiding [2].Ciphering is a method for encoding and manipulating private data so that only authorized users may access or use it [3].The science of information hiding is called steganography [4].It is done by obscuring the existence of the information by blending it into other data [5].A variety of digital mediums, such as image, video, audio, and text, can be utilized to conceal sensitive information [6], [7].The information hiding idea is illustrated in Figure 1.
The vast majority of currently used methods to conceal information rely on image hiding algorithm.This method principally meets the imperceptibility criteria of hiding algorithm by making the little modifications made to the image imperceptible to the human visual system.Some of the image hiding algorithms used are spatial domain method, where the secret information is directly embedded in the pixel intensity, such as least significant bit technique, and another method of the image hiding algorithms is transform domain, where the secret information is concealed in the frequency domain of an earlier transformed image [8].The major problem of data concealing strategies is to include the most information possible while maintaining the image's quality, security, and method resilience to electronic attacks by hackers.Due to the vast number of available digital images on the web and the ease with which the images are used in a hiding process, several schemes have been proposed to hide data in images.Researchers have made an effort to create contemporary strategies to deal with the rapid evolution of hiding techniques [10].
Abdur et al. [11] described a hybrid security method that combines watermarking, steganography, and encryption.It consists of three essential parts: firstly, the secret image is ciphered utilizing the secret key.Secondly, the encrypted image embedded into the cover image utilizing the least significant bit (LSB) method to obtain the stego-image.Thirdly, the ownership of the stego-image has been ensured by using watermarking in spatial and frequency domains, respectively.Their suggested method offers great security against threats and assaults.Lotlikar [12] proposed data hiding method-based cryptography algorithms.The primary goal of their proposed method is to offer a safe means of sending and receiving sensitive and significant data between users in the form of images.
Susanto et al. [13] proposed a method for combination two security techniques, steganography and cryptography.The steganography technique is based on LSB that has benefits in data embedding capability and secret data imperceptibility.Since the LSB approach is so straightforward and predictable, additional encryption techniques are used to increase communication security.The secret data is encrypted using hybrid Beaufort Vigenere (HBV) before embedding it.HBV encryption is a combination of Vigenère cipher and Beaufort algorithm in order to increase data security.
Alabaichi et al. [14] proposed an image steganography utilizing LSB and secret map approaches that performs by applying 3D Chebyshev and 3D logistic maps in order to achieve excellent security.The proposed method is based on the idea of performing random embedding and choosing a pixel from the cover image.The suggested approach is thoroughly assessed using a variety of metrics, including correlation coefficient, homogeneity, information entropy, histogram, concealing capacity, key sensitivity, peak signalto-noise ratio (PSNR), mean square error (MSE), and image fidelity.As a result, their proposed method effectively conceals sensitive information while maintaining a stego-image with high visual quality.
Almayyahi et al. [15] focused on creating a safe technique for concealing information in images using LSB.The Huffman algorithm is used to compress the secret message before it is moved on to the embedding step.Exclusive-NOR (XNOR) operation and the Fibonacci algorithm are then used to choose which pixels to insert secret information.These procedures lead to the creation of a stego-image with two secret keys.The evaluation step of their proposed method demonstrates both the rising security and imperceptibility.
Almazaydeh [8] presented an image steganographic method utilizing a canny edge detector that depends on embedding the private message bits using variable LSB length of one channel from the red, green, and blue (RGB) cover image.The blue channel was chosen because a steganography-based study revealed that blue color intensity is seen visually as being less distinct than red and green colors.Private bits are hidden up to four LSB bits which are chosen using a random number generator.
Ogundokun and Abikoye [16] proposed a modified LSB approach capable of concealing and preserving medical data in order to address the essential authentication problem.Their proposed method employed a logical bit shift operation.Their suggested protected medical information system was shown to be effective in concealing medical data and producing undetected stego-image.An image steganography technique-based Alpha blending is proposed in [17] by utilizing Haar discrete wavelet transform (DWT).DWT is utilized for both the cover image and the secret image.The secret image is ciphered first and then fused to create the stego-image.
Nezami et al. [18] suggested a method that employs the hash function for steganography, and the Caesar cipher and Vigenère to encrypt the message before the hiding process.The proposed approach uses a cover image in the spatial to hide encrypted data.Comparing the suggested method to the conventional LSB steganography methodologies, security and speed are improved.
This paper introduces security procedures for confidential information by combining ciphering and information hiding techniques.Confidential information is ciphered using Caesar cipher with an intelligent method to generate Caesar key by deriving it from the length of confidential information according to a mathematical equation.Thus, the Caesar key exchange between the two ends of the connection is not required.
After that, a 1's complement is taken for each character in the encoding data to provide a second level of security, then the 1's complementary form is hidden using an information hiding technique in the spatial domain with intelligent methods for creating the medium of hiding technique.The hiding medium is created by fusing two images, then scattering the resultant fused image by 1's complement and circularly shifting the bits of fused pixels with specified positions before the hiding process.Thus, the security of the spatial domain pixels is reinforced.
The paper is categorized into the following categories: section 2 introduces the principle of Caesar cipher and fused image.In section 3, details of the proposed algorithm are presented.In section 4, the detailed results were summarized and discussed.Finally, section 5 is considered to be the conclusion of the paper.

PRINCIPLE OF CAESAR CIPHER AND FUSED IMAGE 2.1. Caesar cipher
Caesar cipher is an example of a substitution cipher, where each letter in the plaintext is changed to a letter that is moved a predetermined number of positions down the alphabet.Modular arithmetic is used to illustrate the encryption [19], [20].Julius Caesar, the invention's name, is another name for the Caesar cipher [19].For instance, if the shift was 3, A would become D, B would become E, and C would become F [21] as demonstrated in Figure 2.
One of the Caesar cipher's biggest flaws is that, even in cipher-text only scenarios, it is readily cracked.Various methods that break the text of the codes have been discovered utilizing frequency analysis and pattern words.One method is to employ brute force in order to match the frequency distribution for the letters.Because there are a finite number of shifts that might occur [22].

Fused images
Image fusion is the technique of taking the visually significant information from two or more images and merging them together to create a single fused image [23].It is crucial that the image fusion be carried out without using any sort of final image distortion [24].Figure 3 depicts the wavelet-based image fusion algorithm.The source images are decomposed into approximate coefficient and detailed coefficient at the necessary level utilizing DWT.The fusion rule is used to merge the approximate and detailed coefficients of both source images.By using inverse discrete wavelet transform (IDWT), the fused image is produced [25].

METHOD
The different steps employed to provide security in the proposed method are demonstrated in detail according to the following parts.Part 1 explains the detailed steps that are performed to reinforce the security of the pixels in the spatial domain before the hiding process, while the detailed steps to enhance the security of confidential data are explained in part 2. Finally, the entire proposed schemes at the sending and receiving ends, respectively, are shown in Figures 4 and 5. Part 2: the steps of the security procedure for confidential data before hiding are explained as: a) After reading the secret text, it will be encrypted using Caesar cipher with a new method of formation the Caesar key depending on the length of confidential data according to a mathematical equation.b) The 1's complement for each letter in the cipher data is obtained to provide a second level of security.

RESULTS AND DISCUSSION
The execution of the suggested algorithm is performed by using MATLAB (R2021a).Two grayscale cover images of the same size are required to observe the achievement of the algorithm in offering security for secret information.The database of the signal and image processing institute "SIPI image database" is utilized for the examination.The steps employed for the proposed scheme at the sending part are demonstrated with detailed in the following Figures.Figure 6 illustrates the steps employed for the process of fusing two cover images, the first cover image in Figure 6(a), the first cover image after applying DWT in Figure 6(b), the second cover image in Figure 6(c), the second cover image after applying DWT in Figure 6(d), the process of fusing two cover images in frequency domain in Figure 6(e), and applying IDWT for the fused images in Figure 6(f).As illustrated in Figures 6 and 7, several security steps are performed to reinforce the security of the spatial domain pixels before the hiding process, so the attacker cannot know the pixel values utilized in the hiding process.On the other hand, two security steps are implemented to enhance the security of confidential data before the hiding process, as shown in Figures 8.In addition, the formation of the Caesar key based on the number of characters for the secret data enhances the Caesar cipher secrecy, as the exchanging process of the Caesar key between the sending and receiving parties is not required.
Table 1 shows the Caesar key for the different lengths of confidential data and the time required for the encryption and decryption processes, respectively.According to the Caesar key shown in

CONCLUSION
In this paper, a reliable algorithm for protecting confidential information utilizing fused images is proposed, where four levels of security are provided for the secret information.Firstly, confidential information is encrypted using Caesar cipher with a new method for the formation of the Caesar key.After that, the cipher form of encrypted data is complemented using 1's complement.That procedure enhances the security of Caesar cipher and thus provides a second level of security for secret information.A third level for  ISSN: 2088-8708 Int J Elec & Comp Eng, Vol. 13, No. 6, December 2023: 6797-6805 6804 security is provided using fused two cover images, then scattering the resulting cover image by 1's complement and circularly shifting the bits of fused pixels with specified positions before the hiding process.Finally, the fourth level of security is provided by the hiding technique in the spatial domain, where the complemented form of cipher information is concealed in the scrambled fused image utilizing the LSB technique.Simulation results demonstrate that the proposed approach provides higher security for confidential information.

Figure 4 .Figure 5 .
Figure 4.The steps employed for the sending part

Figure 7 (
a) illustrates the corner replacement of the fused images, 1's complement for the pixels in Figure7(b) and scrambled fused images in Figure7(c).

Figure 8 (
a) illustrates the secret information, cipher of secret information using the Caesar algorithm in Figure8(b), and 1's complement of Caesar cipher in Figure8(c).

Figure 6 .
Figure 6.The steps employed for the process of fusing two cover images (a) 1 st cover image, (b) decomposed 1 st cover image using DWT, (c) 2 nd cover image, (d) decomposed 2 nd cover image using DWT, (e) fused two cover images, and (f) fused images in special domain

Figure 7 .Figure 8 .Figure 9 .
Figure 7.The steps employed for enhancing the security of fused images (a) fused images with corner replacement, (b) 1's complement for the pixels, and (c) scrambled fused images

Table 1 .
The performance of the Caesar cipher with the formation of keyThe number of characters Caesar key Time required for the encryption Time required for the decryption