Digital watermarking is a technique used to safeguard the ownership and integrity of digital material, such as images, sounds and movies, by embedding information or signals. Digital watermarking may serve as a means of verifying the authenticity of data in cloud computing environments, whereby data is stored and accessed remotely. Digital watermarking is a technique that involves the embedding of a distinct watermark onto data to establish and authenticate its legality. Creating a watermark may be achieved by embedding a message or pattern into the data or altering the least significant bits. Digital watermarking techniques enable the verification of medical pictures stored in cloud-based systems. The watermark embedding process involves the insertion of watermarks into medical images using a specific methodology. The selection of an authentication technique is contingent upon the particular characteristics of the protected data and the corresponding security requirements. The medical image, including a watermark, is stored on a cloud server or service [1,2].

The process of obtaining the watermarked medical image: The extraction of the watermark from the medical picture may be achieved by a decoding method after retrieving it from the cloud environment. Verifying the authenticity of a medical picture involves comparing the extracted watermark with the original watermark placed in the image. When the two watermarks are identical, it is deemed that the medical image is actual. In the event of a discrepancy, it might be inferred that the medical picture has undergone manipulation, rendering it unreliable and lacking in accuracy [3,4].

This study contributes substantially to data authentication in cloud settings by focusing on guaranteeing the integrity and authenticity of data stored in distributed and shared cloud-based systems. In situations of this sort, the prevalence of data security issues such as manipulation, theft and loss is attributed to the decentralized data storage structure. The present study employs digital watermarking methodologies to improve data authentication, explicitly focusing on storing and transmitting medical images in cloud environments.

This study presents a revolutionary methodology for including concealed identification data into medical images via digital watermarking techniques while ensuring minor perceptible modification to their aesthetic attributes. The integrated watermark is a crucial component that validates the authenticity and source of medical images. The research progresses through many pivotal phases: Initially, a comprehensive investigation will be conducted on the many aspects of watermarking systems that significantly influence the generation of ideal watermarked data. Additionally, the development of a novel watermarking technique specifically designed for medical images aims to create an optimal trade-off between the watermark's strength and the image's quality, ultimately resulting in an enhanced degree of authenticity.

The efficacy of the suggested methodology is supported by thorough experimentation. The findings emphasize the resilience of the technique and showcase its capacity to preserve picture quality while guaranteeing data integrity and authenticity in a cloud-based setting. Ultimately, this research provides a comprehensive understanding of the functioning of watermarking systems and proposes a pragmatic approach to enhancing the verification process of medical pictures inside cloud-based systems. The use of this approach contributes to the improvement of trust and security in the management of patient data within the healthcare industry.

The present paper has been structured in the following manner: Section 1 of the document included a rudimentary overview of watermarking methods and their use in cloud computing. Section 2 of the composition examines the relevant literature and prior research conducted in the field. Section 3 provides an elucidation of the attributes of cloud computing and the services it offers. Section 4 explains the various forms of digital watermarking. Section 5 presents a comprehensive overview of the procedural framework of the proposed model, while Section 6 provides an in-depth analysis of the obtained outcomes. Lastly, Section 7 offers the concluding remarks.

In 2017, Uma B. and Sumathi S. presented a method to enhance data security for mobile users and the mobile cloud by using an RSA digital signature to encrypt the confidential message the mobile user forms to the mobile cloud through the Internet. Then the information is watermarked using robust reversible watermarking and sent to be stored in the mobile cloud environment. In the decryption and dewatermarking procedure, the embedded image is extracted using a reversible algorithm to remove the secret message, then decrypted and a hash value is tested to ensure its integrity, which leads to an increase in the security of the data [5].

In 2018, S. Thaiyalnayaki and S. Devi suggested a method that depends on various watermarking techniques to improve security and privacy by combining encryption algorithms and different watermarking algorithms to enhance the output messages without losing any information. Consequently, this method increases the speed of developing digital humanities tools [6].

In 2018, Ching-Chun Chang et al. suggested a different paradigm and developed diverse schemes compatible with several types of public-key cryptosystems. The host images are encrypted similarly, with a long sequence of bits considered the input to the encryption process. Also, they proposed encrypting dissimilar bits independently for a fidelity guarantee. The outcomes explain that the error rate approximates zero and achieves state-of-the-art performance [7].

In 2019 Neha Khajanchi and Vishakha Nagrale proposed watermarking technique to protect the copyright in the Cloud environment. They used gray-level co-occurrence matrix and pca algorithm to extract the original images' features and generate the semi-blind watermarking image. The proposed technique contains three stages: Decomposing, embedding and extraction. They used a binary watermark image as a watermark for embedding. The feature of an image is analyzed by the DWT algorithm and with the help of the GLCM algorithm. That result is less complicated and optimal for creating sets of blind watermarks [8].

In 2020, Naseem Shahzad and Mir Aman Sheheryar suggested a system that contracts the extracted data by using transforms based on image watermarking. Then the watermark is embedded in the image by using both the Discrete Cosine Transform (DCT), the DISCRETE FOURIER TRANSFORM (DFT) and the Discrete Wavelet Transform (DWT). The outcomes explain that the watermarked image is solid and secure against signal processing and geometric attacks [9].

In 2022, Alaa Abdulsalam Alarood proposed a watermarking protocol in a cloud environment for buyers and sellers by using a semi-trusted third party for copy prevention and privacy preservation in the cloud computing environment. The proposed protocol uses the Diffie-Hellman key exchange algorithm to offer an encrypted domain for a secure digital media exchange. It adopts a robust watermarking algorithm to protect the image against several attacks. Experiments were done to improve the quality of watermarked digital media, which led to a strong watermark. These experiments were done to see how well the suggested protocol worked against the attacks [10].

In 2022, Namita D. Pulgam and Subhash K. Shinde proposed a method that depends on the value of image intensity by using pixel color correlation (WPCC) and chaos-based encryption to use the Arnold transformation, which creates many levels to protect the medical record of the patient and then embeds it as a watermark in the medical image. The suggested method produces images with a good Peak Signal Noise Ratio (PSNR) that ranges between 24.74 dB and 36.07 dB.z Also, the structural similarity index (SSIM) ranged between 0.84 and 0.97. They checked and evaluated the outcomes by using the Bit Error Rate (BER) and Normalized Correlation Parameter (NCC) and found that the proposed method is capable of extracting and hiding the medical record of the patient securely and firmly against several attacks [11].

Cloud Computing 

In the present year, cloud computing can be considered one  of  information  technology's  most  crucial  computing models. Cloud computing delivers computing services over the Internet, allowing users to access a shared pool of computing resources such as servers, storage, databases, networking, software and applications on demand. Instead of hosting and managing their IT infrastructure, users can use the resources and services of cloud providers to scale up or down as needed, pay for only what they use and avoid the costs and complexity of maintaining their hardware and software [12]. There are several definitions of cloud computing but NIST produces the final report of cloud computing as a "model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud comprises five essential characteristics, three service models and four deployment models." [13]. Cloud computing enables clients to run applications and access data from any device, place and time without uploading or upgrading the applications on local machines or servers. This is one of the most essential advantages of cloud computing, making it the most common computing model [14]. Cloud computing is often categorized into three main types of services: Infrastructure as a Service (IaaS) provides virtualized computing resources, such as servers, storage and networking, that users can rent and configure to create their IT environments. Platform as a Service (PaaS) provides a platform for developing, running and managing applications without controlling the underlying infrastructure. Software as a Service (SaaS) provides applications and software that users can access and use over the Internet without the need to install or manage any software on their own devices [15]. The cloud has the five essential characteristics Necessary to deploy cloud services rapidly and cost-effectively. These characteristics are [16]:

• Self-Service on demand: clients can demand and supply computing capabilities with any services without going through an admissions process powered by automation or workflows

• Ubiquitous Access to Network: Clients and computing devices can be accessed via the network even using heterogeneous devices of different generations, like smartphones, tablets and thick or thin clients

• Resource Pooling: The data centers are divided into several pools that provide services to diversified clients in a multi-tenant manner. These pools can be both virtual and physical resources

• Rapid Elasticity: That makes the supply elastic and fast. This supply can be done automatically. The client can use the unlimited capacity as a service that can be bought anytime

• Measured Service: To optimize the cloud environment, it must manage the workload effectively. The management requires measuring the service, metering, monitoring and improvement at the user level

Because of the improvement in technology, a large quantity of data is stored on cloud servers, which becomes an aim for many hackers. When these data are exposed, the damage to society and clients will be significant. The vulnerabilities of stored personal data, trading assets, health information and intellectual rights will lead to great destruction. Therefore, it is necessary to use authentication methods to protect personal data or information so that only authorized clients can access it [17].

Digital Watermarking

Digital watermarking is the technique of embedding data into digital multimedia content. It is used to confirm the credibility of the content or to identify the identity of the digital content owner. Digital watermarking can be divided into four types depending on the kind of document to be watermarked: text watermarking, audio watermarking, image watermarking and video watermarking [18]. The digital watermarking method generally consists of two main stages: watermark embedding, which embeds the watermark in the cover or original data using the embedding functions and the watermarking key. The watermark extraction was used later to detect and extract the watermark signal from the watermarked data [19]. A watermarking key is used in the embedding and extraction process to avoid unauthorized access to the watermark [10,21].

Type of Digital Watermarking

Watermarking is a branch of information hiding that can be considered one of the most important ways to protect digital content. The increased use of cloud computing in several transactions and services led to the emergence of different types of watermarking techniques to deal with varying types of data [18]. These types include:

• Visible Watermarking: This type is visual, which means it can appear on the surface of the multimedia. It can be logos or text put over the digital media, such as branding determination [22]

• Invisible Watermarking: It is embedded in the digital media but not visible and only particular software detection can decode and detect it [23]

• Fragile Watermarking: It exposes any manipulation in digital media. It is added to digital media and altered or destroyed if any changes are happened [24]

• Robust Watermarking: protects digital media against unauthorized copying, use and distribution. It is unaffected by compression, filtering, cropping and attacks [10]

• Semi-fragile Watermarking: it is a mixture of robust and fragile watermarking. It is used to discover any intended or unintended alteration in the digital media and authenticates the content or detects tampering [24]

The Proposed Model

Digital watermarking for medical images embeds hidden information in the images themselves. This embedded information, known as a watermark, can serve various purposes,     including     image     authentication,    copyright protection and data integrity verification. The proposed system employs the Least Significant Bit (LSB) technique as one of the most common spatial domain techniques. First, use Otsu's segmentation method to preprocess the input medical image and divide it into two parts, Region of Interest (ROI) and Non-Region of Interest (NROI). ROI is ignored because it contains essential details, while NROI is binarized using the LSB technique. Then replace the rightmost bits of every pixel with the watermark bits. The watermark is generated by encoding the name and birthday date of the patient into an ASCII code vector. To enhance the system's security, encryption techniques can be used to protect the watermark before embedding it into medical images for increased safety using Advanced Encryption Standard (AES). The proposed method can be summarized in the following stapes as shown in Figure 1:

• Enter patient information such as name and birthday date

• Prepare the watermark that will be embedded into the patient's medical image

• Encrypt the watermark using Advanced Encryption Standard (AES) algorithm

• Get the medical image of the patient and perform Otsu's method

• Extract the Region of Interest (ROI) and Non-Region of Interest (NROI) for the medical image

• Embed the watermark: Implement the selected algorithm to embed the 64-bit of the watermarking vector into (the NROI) of the medical image. This involves modifying the pixels to incorporate the Least Significant Bit (LSB) watermark while preserving the content's quality and integrity

Figure 1: The Proposed Model

The watermarked medical images are then stored or transmitted to the cloud environment. The cloud can be a private cloud for healthcare institutions or a public cloud for secure data storage and sharing.

The watermark extraction process is used to recover the embedded water from the watermarked medical picture, using a specific key to retrieve the embedded watermark successfully. Subsequently, the researchers used both the designated key and the stated identification watermark during the detection procedure to ascertain the  accuracy  of  the  watermark.  Upon  the  authorized user retrieves the medical picture from the cloud, a process is initiated to verify the image's validity. The presence of disparities or the absence of watermarks would suggest that the image has been manipulated or lacks authenticity.

Experimental Results

The  proposed  system  achieves  a  high  performance over  many  medical  images  with  different  sizes  for CT  Medical  Images.  Figure 2(A-C)  explain  the difference between the original and watermarked medical images and the preprocessed medical images between them. LSB watermark algorithm contributes to  the  rapidity  of  the  system  and  reduces  its complexity. In addition, the watermarked image's security is increased due to using the encrypted patient information as a watermark.

Figure 2A shows the original medical images that have been used for watermark embedding, while Figure 2B shows the corresponding preprocessed medical images that have been segmented into two parts (ROI and NROI). Figure 2C shows the watermarked medical images. It is challenging to notice any effect of the hidden watermark inside the medical images. To assess the performance of the proposed method, we used both Peak Signal-to-Noise Ratio (PSNR) and Mean Structure Similarity Index (MSSIM). That evaluates the distortion and security measures for watermarked medical images compared to the original ones. Table 1 shows the different test outcomes for five diverse images.

Table 1: Assessment Results for the Proposed Technique

Figure 2(A-C): (A) Original Images, (B) Preprocessed Images and (C) Watermarked Images.

In this paper, we developed a medical image watermarking approach to protect the patient's medical image before sending it to the cloud. Different sophisticated image processing tools can alter medical images. Therefore, medical images must be protected from unauthorized access for secure communication. In this approach, we use patient information to create the watermark, then encode the watermark using the AES algorithm to improve the security level of the watermarked medical image. Otsu's method was used to determine the region that does not contain any critical information to embed the watermark inside. That leads to more authentication and integrity verification.

1. Garg, P. and R. Kishore. "Performance Comparison of Various Watermarking Techniques." Multimedia Tools and Applications, 2020, pp. 1-47. https://doi.org/10.1007/s1 1042-020-09262-1.

2. Anand, A. and A. Singh. "Dual Watermarking for Security of COVID-19 Patient Record." IEEE Transactions on Dependable and Secure Computing, vol. 20, 2023, pp. 859-866. https://doi.org/10.1109/TDSC.2022.3144657.

3. Liu, B. et al. "Privacy Protection for 3D Point Cloud Classification Based on an Optical Chaotic Encryption Scheme." Optics Express, vol. 31, no. 5, 2023, pp. 8820-8843. https://doi.org/10.1364/oe.483522.

4. Harika, D. and S. Noorullah. "Implementation of Image Authentication Using Digital Watermarking with Biometric." International Journal of Engineering Technology and Management Sciences, 2023. https://doi.org/10.46 647/ijetms.2023.v07i01.023.

5. Uma, B. and S. Sumathi. "An Efficient Approach for Data Security in Cloud Environment Using Watermarking Technique and RSA Digital Signatures." International Research Journal of Engineering and Technology (IRJET), vol. 4, no. 2, 2017.

6. Thaiyalnayaki, S. and S. Devi. "Protection of Data in Cloud Computing Using Image Processing - Watermarking Technique." International Journal of Computer Sciences and Engineering, vol. 6, no. 11, 2018, pp. 115-118.

7. Chang, C.C. et al. "Privacy-Aware Reversible Watermarking in Cloud Computing Environments." IEEE Access, vol. 6, 2018, pp. 70720-70733.

8. Khajanchi, N. and V. Nagrale. "To Apply Watermarking Technique in Cloud Computing to Enhance Cloud Data Security." International Journal of Scientific Development and Research (IJSDR), vol. 4, no. 7, 2019.

9. Shaizad, N. and M.A. Sheheryar. "Watermarking Technique to Enhance Image Security in Cloud Computing." International Journal of Management, IT & Engineering, vol. 10, no. 8, 2020.

10. Alarood, A.A. "Improve the Efficiency for Embedding in LSB Method Based Digital Image Watermarking." Journal of Theoretical and Applied Information Technology, vol. 100, no. 15, 2022.

11. Pulgam, N.D. and S.K. Shinde. "Robust Digital Watermarking Using Pixel Color Correlation and Chaotic Encryption for Medical Image Protection." International Journal of Intelligent Systems and Applications in Engineering, vol. 10, no. 4, 2022, pp. 29-38. https://ijisae.org/index.php/IJISAE/ article/view/2193.

12. Kumar, S. et al. "Digital Watermarking-Based Cryptosystem for Cloud Resource Provisioning." International Journal of Cloud Applications and Computing, vol. 12, no. 1, 2022.

13. Boussif, M. et al. "Secured Cloud Computing for Medical Data Based on Watermarking and Encryption." The Institution of Engineering and Technology (IET) Networks, 2018.

14. Ouda, G.K. "Cloud Computing Service Providers: A Comparative Study." Samarra Journal of Pure and Applied Science, 2020.

15. Ara, R. et al. "Cloud Computing: Architecture, Services, Deployment Models, Storage, Benefits and Challenges." International Journal of Trend in Scientific Research and Development (IJTSRD), vol. 4, no. 4, 2020.

16. Kaur, H. "Characteristics of Cloud Computing." International Journal of Research and Analytical Reviews (IJRAR), vol. 7, no. 1, 2020.

17. Bollinadi, M. "Cloud Computing: Security Issues and Research Challenges." Journal of Network Communications and Emerging Technologies (JNCET), vol. 7, no. 11, 2017.

18. Khajanchi, N. and V. Nagrale. "To Apply Watermarking Technique in Cloud Computing to Enhance Cloud Data Security." Journal of Emerging Technologies and Innovative Research (JETIR), vol. 6, no. 6, 2019.

19. Gaata, M.T. and M.D. Al-Hassani. "Underwater Image Copyright Protection Using Robust Watermarking Technique." Indonesian Journal of Electrical Engineering and Computer Science, vol. 29, no. 2, 2023.

20. Hatoum, M. Digital Watermarking for PDF Documents and Images: Security, Robustness and AI-Based Attack. Doctoral thesis, UBFC University, France, 2020.

21. Sattar, I.A. and M. Gaata. "Image Steganography Technique Based on Adaptive Random Key Generator with Suitable Cover Selection." Annual Conference on New Trends in Information & Communications Technology Applications (NTICT), 2017.

22. Embaby, A.A. et al. "Digital Watermarking Properties, Classification, and Techniques." International Journal of Engineering and Advanced Technology (IJEAT), vol. 9, no. 3, 2020.

23. Qureshi, A. et al. "Detecting Deepfake Videos Using Digital Watermarking." APSIPA Annual Summit and Conference, 2021.

24. Liu, K. et al. "A Cloud-User Protocol Based on Ciphertext Watermarking Technology." Security and Communication Networks, 2017.