Advanced Dual Reversible Data Hiding: A Focus on Modification Direction and Enhanced Least Significant Bit (LSB) Approaches

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, the author proposed a Reversible Data Hiding algorithm. Equipped with an Exploiting Modification Direction strategy, they improve the data hiding by modifying pixel pairs. Moreover, they minimize the errors of the marked images by an inherent loss function. To evaluate the performance of their method, they compare the method with other baselines. The topic is quite innovative. I suggest the author to make a minor revision. After reading this paper, I have following suggestions:

1^st  In section 3, the author's claim that their method incorporates pair-wise information and utilizes the LSB replacement strategy to hide data lacks a clear introduction of the benefits. To address this, I suggest the author includes examples to demonstrate the effectiveness of their proposed network.

 

2^nd  In 3.2, the author introduce the image recovering strategy and give a briefly introduction in section 3.3. However, they fail to clearly explain the procedure. I hope they can clarify.

 

3^rd  I suggest them to add a pipline framework plot to show the whole procedure of the proposed method.

 

4th I suggest them to discuss the robustness of their method toward data with different level of noise. Also, if possible, they could discuss the limitation of their work.

 

5th The writing needs to be further polished.

 

 

Comments on the Quality of English Language

need to polish

Author Response

Dear Reviewer1

Thank you for reviewing our paper during your busy schedule. We will respond to comments on the paper as follows.

(1) In section 3, the author's claim that their method incorporates pair-wise information and utilizes the LSB replacement strategy to hide data lacks a clear introduction of the benefits. To address this, I suggest the author includes examples to demonstrate the effectiveness of their proposed network.

Response: In response to the reviewer's comment requesting a clearer introduction of the benefits of our method that incorporates pair-wise information and utilizes the LSB replacement strategy for data hiding, we would like to elaborate on the effectiveness and efficiency of our proposed system with reference to Figure 3 provided in our paper.

Our method's primary advantage is its enhanced security and integrity in data hiding. The use of a pair-wise approach allows for the embedding of a quinary digit across two pixels, which means that we can effectively hide approximately 2.32 bits of information (as calculated by log₂(5)) in a single operation. When this is applied across the image, our method achieves an impressive data hiding efficiency of approximately 1.16 bits per pixel pair.

Specifically, our approach minimally alters the pixel values, with a change of at most plus or minus one, ensuring that the visual integrity of the cover images remains intact. This subtle alteration, which is practically undetectable to the human eye, serves to maintain the cover images' aesthetic quality while also securing the embedded data against unauthorized detection and extraction.

Furthermore, the receiver can successfully extract the hidden data and recover the original image using the pair of stego images, as shown in Figure 3. This two-fold process—data hiding and image restoration—is seamless, further showcasing the robustness of our method.

In summary, our method does not compromise on image quality for the sake of data concealment and provides a clear path for both the concealment and subsequent recovery of data, offering a more secure and efficient solution compared to traditional single-image methods. We believe the inclusion of specific examples and illustrations of this process, as seen in Figure 3, will aid in comprehending the full capabilities and advantages of our proposed system.

(2) In 3.2, the author introduces the image recovering strategy and give a briefly introduction in section 3.3. However, they fail to clearly explain the procedure. I hope they can clarify.

Response: Thank you very much for your valuable feedback regarding sections 3.2 and 3.3 of our manuscript. We appreciate the opportunity to clarify the procedures related to data extraction and recovery, as well as our approach to managing underflow and overflow scenarios in the embedding phase of our steganographic method.

In response to your comments:

1. Section 3.2: We have revised the section to include a step-by-step description of the data extraction and recovery process. This includes how we handle pixel pairs during the extraction phase to ensure the accurate retrieval of hidden information and the restoration of the original image. We have outlined the mathematical operations and logical conditions applied to each pixel pair to retrieve the embedded data and reconstruct the pixel values of the original image.
2. Section 3.3: We realized that more explicit details could enhance the reader's understanding. Therefore, we have expanded this section to include examples of how pixel values at the extremities are managed. Specifically, we described the precautionary exclusion of pixel pairs with values (0,0) and (255,255) from the embedding process, and how alternative pixel pairs are selected to maintain the integrity of the embedded data and prevent potential errors during data extraction.

We believe these revisions will significantly improve the clarity of the described procedures and the overall quality of the manuscript. We are committed to ensuring that our work is presented as clearly and comprehensively as possible, and we hope that these changes adequately address your concerns.

 

(3) I suggest them to add a pipeline framework plot to show the whole procedure of the proposed method.

Response: Thank you for your valuable feedback on our manuscript, specifically regarding Figure 3. We have taken your comments into consideration and have updated the figure to enhance its clarity and comprehensiveness. In the original version of Figure 3, we presented a diagram that solely depicted the sender’s perspective in the data hiding process. It detailed how two cover images are generated from the original image and then utilized in conjunction with our data hiding algorithm. However, upon your suggestion, we recognized the necessity to illustrate the complete process. Therefore, in the revised version of the diagram, we have expanded the scope to include the receiver’s side. This updated diagram now demonstrates how the receiver obtains the two cover images and applies the data extraction and image recovery algorithm to retrieve the hidden data and reconstruct the original image.

 

(4) I suggest them to discuss the robustness of their method toward data with different level of noise. Also, if possible, they could discuss the limitation of their work.

Response: Thank you very much for your insightful comments and suggestions regarding our manuscript. We greatly appreciate your efforts to thoroughly review our work and provide constructive feedback.

Regarding your suggestion to discuss the robustness of our method toward data with different levels of noise and to address the limitations of our work, we would like to clarify the scope and primary objectives of our research within the context of data hiding technologies.

Our study focuses on developing and evaluating a data hiding method, with a primary emphasis on the security and detectability aspects of the hidden data. In the realm of data hiding research, the main goal is often to ensure that the hidden information remains secure and undetectable, rather than to extensively simulate or verify the system's robustness against noise or other forms of distortion. This distinction is particularly relevant as the robustness against external manipulations, such as noise addition, is more commonly associated with watermarking research, which has different objectives and evaluation criteria.

We acknowledge the importance of robustness in multimedia security technologies and understand that evaluating the system's performance under various noise conditions could add value to the research. However, given the specific focus and objectives of our current study, we believe that an in-depth exploration of noise resilience falls outside the immediate scope of our work. Our primary aim was to explore the efficiency and effectiveness of the proposed data hiding technique, rather than its resistance to noise or distortion.

That said, we recognize the significance of your suggestion and agree that future research could benefit from a more comprehensive examination of these aspects. Exploring the robustness of data hiding techniques against various levels of noise and understanding the limitations of these methods in more detail could indeed provide valuable insights and contribute to the advancement of the field. We are grateful for your recommendation and will consider investigating these dimensions in our future work.

We hope this clarifies our position and the scope of our research. Once again, we thank you for your valuable feedback and for providing us with the opportunity to further refine our work.

 

(5) The writing needs to be further polished.

Response:

We deeply appreciate your valuable feedback. Based on your suggestion that "the writing needs further polishing," we have dedicated additional efforts to enhance the clarity and expressiveness of our manuscript.

To improve the readability and impact of our paper, we have taken the following actions in close collaboration with professional editors or fellow researchers:

1. Language and Grammar Review: We meticulously reviewed the language and grammar, ensuring that the expressions used were clear and consistent.
2. Structural Improvements: We enhanced the flow and structure of the paper, adjusting the logical connections between sections to facilitate reader comprehension.
3. Clarification of Terms and Definitions: We clearly explained specialized terms and definitions, making them accessible not only to experts but also to general readers.

 

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper is about the proposal of a reversible data hiding method in
images, with a balance between high data embedding rates and the
integrity of visual quality.

In the title, the acronym LSB could be replaced by the unabbreviated
expression.

1. Introduction

At the beginning of the introduction, it should be said why it is
desirable for the data embedding to be reversible in terms of the
recovery of the image, given that it is only a vehicle for the
message.

In page 2, line 52, it is written

"the k secret bits are replaced by the k least significant bits of
each cover pixel"

Wouldn't

"the k secret bits replaces the k least significant bits of each cover
pixel"

be the correct sentence?

The introduction presents well the techniques previously developed for
data embedding in images.

2. Related Works

The section presents details of existing techniques well.

3. Proposed method

In page 8, line 258, it is written

"it is assumed that the order information of two stego images is
specified in the header of the image".

How is this done?

In the mechanism for using the proposed technique, would sending a
secret message require sending two images? It would be good to clarify
this.

Other than that, the section explains the proposed method well.

4. Experimental results

The experimental results were presented comprehensively and supported
the conclusion that the proposed method is a contribution to the state
of the art.

5. Conclusions

The conclusions are appropriate and compatible with the previous
sections.

The list of references is comprehensive, covers the fundamentals used
and has recent works.

Comments on the Quality of English Language

In page 2, line 41, repetition: "the the".

Author Response

Dear Reviewer2

Thank you for reviewing our paper during your busy schedule. We will respond to comments on the paper as follows.

(1) In the title, the acronym LSB could be replaced by the unabbreviated expression.
Response: The title of the paper was revised as follows.

TITLE: Advanced Dual Reversible Data Hiding: A Focus on Modification Direction and Enhanced LSB Approaches

(2) At the beginning of the introduction, it should be said why it is desirable for the data embedding to be reversible in terms of the recovery of the image, given that it is only a vehicle for the message.

Response: The advantages of RDH are as follows:

“Reversible Data Hiding (RDH)[5-8] technology plays a crucial role in achieving information security and data integrity simultaneously. The core advantage of RDH lies in its ability to restore the original cover image perfectly to its original state after data extraction. This means that RDH can enhance information security while maintaining the quality of the cover image, without losing any original data during the data hiding process. RDH offers a high level of security while ensuring the integrity of data and images. This characteristic makes RDH particularly useful in fields that require high accuracy and protection, such as legal documents, medical images, and copyrighted content. RDH's unique position in the data hiding technology spectrum highlights its value in applications where both the complete recovery of original data and the necessity for security are maximized.”

(3) Wouldn't "the k secret bits replaces the k least significant bits of each cover pixel" be the correct sentence?

Response: The reviewer's advice is very sound, and I have modified it as follows:
“the k secret bits replace the k least significant bits of each cover pixel, thereby embedding the hidden data into the image.”

(4) In page 8, line 258, it is written "it is assumed that the order information of two stego images is specified in the header of the image".

Response: Every image has a header that contains the image's characteristics. Therefore, the following content was added to this paper to prevent readers from having any questions about the order of the images.

 “The header of each stego image contains metadata pertaining to the image. This metadata encompasses not only basic details like the image's size, format, and color depth but also supports the inclusion of additional information for specific application requirements. In this study, the sequence information of stego images is captured as an extra field within the image header's metadata, ensuring these images are processed and interpreted in their intended order. Consequently, algorithms or systems utilizing stego images can consult the image header to determine the sequence of images, which is critical for accurately performing data extraction and restoration processes.”

(5) In the mechanism for using the proposed technique, would sending asecret message require sending two images? It would be good to clarify this.

Response: The reason why the method proposed in this paper requires two cover images is explained as follows. This content has been added to the main text.

 

“Contrary to traditional EMD and LSB-based methods that depend on a single cover image—resulting in the original image being non-restorable—our novel approach employs two cover images. This dual-image strategy not only facilitates more effective data concealment but also enables the successful extraction and restoration of the original images by the receiver, providing a robust solution for secure data management. This method introduces significant security and data integrity advantages. It necessitates the combination of two images for the complete recovery of hidden data, inherently dispersing security risks that come with the compromise of a single image. Such an approach ensures the irrecoverability of data from a single image, thus bolstering the overall security of the communication process. Moreover, our method is designed to maintain the cover image's quality during the data embedding phase. By altering only one pixel in each pair, the method minimally impacts the cover image, ensuring that the modifications remain undetectable and the natural appearance of the image is preserved. The adoption of pair-wise data hiding alongside a dual-image strategy significantly enhances the security and integrity of the hidden data. This not only strengthens the steganographic communication's robustness but also maintains the aesthetic quality of the cover image, offering an effective and secure data hiding solution that meets complex data protection requirements.”

Author Response File: Author Response.pdf