Dual Reversible Data Hiding Based on AMBTC Using Hamming Code and LSB Replacement
Round 1
Reviewer 1 Report
1. State of art should be given clear in the introduction part
2. Author has to highlight the research gap in literature survey, how the previous works carried out and performance of the existing methods
3. Author has to highlight the scope of the research work and mentioned the importance of their research
Author Response
We would like to thank you for reviewing the paper, and the answers to your questions are as follows.
- State of art should be given clear in the introduction part. 1.
Response: The reviewer asked what the state of art in this paper is. Our proposed method is the first to proposed dual RDH based on AMBTC and there is an advantage that the receiving side can restore the cover image. AMBTC is a method that can compress a natural image. It has simple compression calculations and relatively good compression efficiency, so it is very efficient for image and video transmission in IoT-based systems. For data hiding, we propose a Hamming code and LSB replacement method. As a result of the experiment, data hiding, extraction, and cover image restoration were shown, and the excellent performance was verified through the experiment. Response to question is mentioned in the introduction.
- Author has to highlight the research gap in literature survey, how the previous works carried out and performance of the existing methods.
Response: In this paper, we briefly summarize the representative papers related to dual RDH. Our proposed method is dual RDH based on AMBTC using Hamming code and LSB replacement. There have been studies on gray image-based dual RDH, but there is no AMBTC-based study. For this reason, direct comparison with existing studies is not meaningful. Using Table 3 and Table 4 in Section 4, we showed the performance of our proposed method. If our answers are not sufficient, please ask again.
- Author has to highlight the scope of the research work and mentioned the importance of their research.
Response: The research scope of this paper is briefly shown using Figure 1. Most of the existing papers are based on gray images, but the difference may be that this paper is based on AMBTC. Since the characteristics of the image are different, the method of hiding the data is inevitably different. The importance of data hiding is the same as the existing data hiding method. That is, it is used to secretly deliver confidential information to the recipient. Alternatively, it can be used for copyright protection for images and videos. The contribution of our proposed paper is shown in the introduction, so please refer to it.
The following sentence is added at the end of the introduction.
Our proposed method is the first to proposed dual RDH based on AMBTC. AMBTC is a method that can compress a natural image. It has simple compression calculations and relatively good compression efficiency, so it is very efficient for image and video transmission in IoT-based systems. The importance issue of RDH is to secretly deliver confidential information to the recipient. Moreover, it can be used for copyright protection for images and videos.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors of this work introduce AMBTC-based dual-image RDH for the first time and provides high ER and PSNR. The following points have been identified by the present reviewer:
1. The discussion regarding cryptography is superficial, it does not reflect the current state of classical cryptography or emerging techniques (quantum cryptography).
2. The main problem with this approach is that, at its destination, two images must arrive, identical in appearance, which could alarm the attacker, who has copies of both images, therefore proceeds to analyze the images to extract the secret information.
3. The second disadvantage of this approach is that the data throughput is doubled because two images must be sent instead of a single image.
4. A detailed comparison is not presented with respect to other current works in the area to highlight advantages that justify this proposal.
Therefore, the article is not accepted for publication.
Author Response
We would like to thank you for reviewing the paper, and the answers to your questions are as follows.
- The discussion regarding cryptography is superficial, it does not reflect the current state of classical cryptography or emerging techniques (quantum cryptography).
Response: This paper is not about cryptography. Data hiding is a method for hiding confidential information in an image and delivering it safely to the recipient. This method can be used for copyright protection of images, etc.
- The main problem with this approach is that, at its destination, two images must arrive, identical in appearance, which could alarm the attacker, who has copies of both images, therefore proceeds to analyze the images to extract the secret information.
Response: If a third party steals the two images, they can't extract the data because they don't know how to decode it. The receiver can extract data by using a decoding method after collecting two images, and it is difficult to know the contents of the message because secret information cannot be extracted with only a single image. This method is not generally a research paper based on an encryption protocol. That's why we don't cover protocols for defending against images taken over by third-party attackers. This method is a study focused on how safely it can hide data and evade attackers' eyes when hiding data in an image.
- The second disadvantage of this approach is that the data throughput is doubled because two images must be sent instead of a single image.
Response: Compared to data hiding for a single image, data throughput may be slightly longer when using two images. However, encoding does not require real-time processing. Also, in this paper, two identical cover image-based data hiding methods use some of the advantages of the existing secret sharing method, and secret sharing improves security by itself. This is an advantage of dual image-based data hiding. Also, the fact that the amount of data hiding can be increased can be an advantage.
- A detailed comparison is not presented with respect to other current works in the area to highlight advantages that justify this proposal.
Response: Our proposed method is dual RDH based on AMBTC using Hamming code and LSB replacement. There have been studies on gray image-based dual RDH, but there is no AMBTC-based study. For this reason, direct comparison with existing studies is not meaningful. Using Table 3 and Table 4 in Section 4, we showed the performance of our proposed method. If our answers are not sufficient, please ask again.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper is well written, is clear and can be well understood. It is relevant and can be of help and interest for the information security research community.
The topic is suitable for the journal, of broad international interest, significant and novel. The paper is clearly presented. The references are relevant, up to date, accessible, adequate and complete.
In this paper, a new data hiding is a method of secretly transmitting a message using an existing channel after hiding data in one cover image. The authors propose a Absolute Moment Block Truncation Coding based dual-image Reversible Data Hiding using Hamming Code. The proposed method introduces AMBTC-based dual Reversible Data Hiding for the first time and provides a system with better Image Quality Metrics as embedding ratio and PSNR than previous existing systems. Because dual Reversible Data Hiding is used, there may be improvement in terms of security compared to using a single cover image. The proposed method has the advantage that the quality of the two marked images is almost the same because the payload is properly distributed. The paper present ten experimental examples, with images from the standard USC-SIPI image database, the results show the good behavior of the proposed chaotic image encryption algorithm.
The work is enough relevant to be published by the journal Electronics, because the interest of the chaotic information encryption methods is very high for the security research environment.
Author Response
We would like to thank you for reviewing the paper.
Round 2
Reviewer 1 Report
All the comments are properly addressed.
Check English Grammar.
Author Response
We checked the English grammar and made some corrections.
Reviewer 2 Report
This reviewer perceives as not addressed, or not addressed in depth and with due argumentation, the observations identified in the first round. For this reason, acceptance to publish this manuscript is not considered. Briefly, the detected points are listed again:
1. The discussion regarding cryptography is superficial.
2. At its destination, two images must arrive, identical in appearance, which could alarm the attacker, who has copies of both images, therefore proceeds to analyze the images to extract the secret information.
3. The data throughput is doubled because two images must be sent instead of a single image.
4. A detailed comparison is not presented with respect to other current works in the area to highlight advantages that justify this proposal.
Author Response
Thank you very much for taking the time to review the paper. However, we mentioned in round 1 that the important issue of this paper is not cryptography or cryptography protocols. Again, this paper is not related to cryptography or cryptography protocols. We would be very grateful if you could understand this and review the paper.
1. The discussion regarding cryptography is superficial.
>>> Since this paper is not related to cryptography or cryptography protocols, this part is not covered. The realm of data hiding does not consider the issues mentioned.
2. At its destination, two images must arrive, identical in appearance, which could alarm the attacker, who has copies of both images, therefore proceeds to analyze the images to extract the secret information.
>>> The mentioned parts are not considered as issues of this paper. The realm of data hiding does not consider the issues mentioned.
3. The data throughput is doubled because two images must be sent instead of a single image.
>>> The throughput may be doubled rather than one. However, since this paper is not relevant to the real-time processing research area, the issues raised cannot be considered.
4. A detailed comparison is not presented with respect to other current works in the area to highlight advantages that justify this proposal.
>>> We mentioned the strengths of the paper in the introduction. We also mentioned performance in Section 4.
Author Response File: Author Response.pdf
