Remote Sensing Scene Graph and Knowledge Graph Matching with Parallel Walking Algorithm
Round 1
Reviewer 1 Report
Dear Authors,
The revision of the article has been done, and some suggestions were made in the article.
The following aspects are brought to the attention of the authors.
1.
Abstract
The abstract is much too consistent, in relation to the Instructions for Authors, and Microsoft Word template, Remote Sensing journal.
It could be revised.
2.
Space before citing a bibliographic source in the text
eg
Page 1, row 44
“model [1-3]” instead of “model[1-3]”
Several suggestions were made in the article
3.
Article structure
It is recommended that the structure of the article corresponds to the Instructions for Authors, and Microsoft Word template, Remote Sensing journal.
“Research Manuscript Sections”
Introduction
Materials and Methods
Results
Discussion
Conclusions
It is possible that by rearranging the content according to the recommended chapters structure, the order of citation of some bibliographic sources may undergo changes.
Revision is recommended to be consistent with their order of presentation in the References chapter.
In the Discussion chapter, it is recommended to discuss the results obtained in relation to other studies in the field.
The last bibliographic reference cited "[41]" is presented in the article on page 6, row 265.
The bibliographic source "[42]" (page 15, row 575) is not presented in the References chapter
For Discussions, additional bibliographic sources can be added.
Revision is necessary
4.
A sentence cannot begin with a parenthesis.
eg
Page 2, row 97
"Zhang et al. [11]" instead of "[11]"
Similarly for other bibliographic sources
e.g
page 3, rows 102, 108, 112, 116, 126, 131,
page 4, rows 155, 159, 162, 173, 176, 195, 196,
5.
Text setting
It is recommended to follow the recommendations for text settings
Styles: MDPI_3.1_text
Eg
Page 5, rows 217 – 220.
6.
Tables titles setting
It is recommended to check the table title settings
Styles: MDPI_4.1_table_caption
7.
Author Contributions:
Needs revision in accordance with Instructions for Authors, and Microsoft Word template, Remote Sensing journal
eg
“The following statements should be used “Conceptualization, X.X. and Y.Y.; methodology, X.X.;”
8.
References
The References chapter needs to be revised, according to the Instructions for Authors, Remote Sensing journal.
“Author 1, A.B.; Author 2, C.D. Title of the article. Abbreviated Journal Name Year, Volume, page range.”
eg
“Cui, W.; He, X.; Yao, M.; Wang, Z.; Hao, Y.; Li, J.; Wu, W.; Zhao, H.; Xia, C.; Li, J.; Cui, W. Knowledge and Spatial Pyramid Distance-Based Gated Graph Attention Network for Remote Sensing Semantic Segmentation. Remote Sens. 2021, 13(7), 1312. https://doi.org/10.3390/ rs13071312 “
Instead of
“Cui W , He X , Yao M , et al. Knowledge and Spatial Pyramid Distance-Based Gated Graph Attention Network for Remote Sensing Semantic Segmentation[J]. Remote Sensing, 2021, 13(7):1312.”
Comments for author File: Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
The current article proposes a scene graph optimization module based on cosine similarity graph pooling. The proposed method aims to achieve graph matching between generalized knowledge and specific image scene. Also, it can integrate geographic knowledge to identify complex geographic scenes. Moreover, it constructs and optimizes the sample scene graph and provides a more precise search space for graph matching. The method is also equipped with a scene graph-knowledge graph matching module based on the parallel walking algorithm for accurate and efficient graph matching. Finally, the knowledge sub-graph-based super-pixel group classification module is designed to improve the recognition of complex geographical scenes.
Unfortunately, the paper contains the following problems.
• The paper language needs polishing since it contains grammar and syntax errors.
• The abstract is too long and must be rewritten more compactly.
• The authors introduce abbreviated terms without writing the full terms the first time they are introduced (e.g., what is KGGAT in the abstract. The text contains more terms that need addressing).
• The related works inside the documents are not appropriately referenced (e.g., in line 126 [17] proposed... must change to Chen Peng [17] proposed… The article contains more references that need addressing).
• In the related work part of the article, the “Summary:” terms must be removed, and the authors must replace them with a comparison of their proposed method with the existing works.
• All figures inside the text must contain a small paragraph explaining them. This paragraph must be placed next to each figure’s caption.
• Figures 1-3 are too small and difficult to read.
• In sub-section 3.1, the authors introduced four modules that are part of their architecture without first explaining how they are connected. Before explaining these modules, the authors must add a paragraph explaining figure 1, which contains the structure of PWGM.
• The figure caption on page 6 must be right under the figure and not on a separate page.
• The sub-section titles must be together with at least a small part of their text and not alone on a separate page (e.g., the Graph Pooling sub-section title on page 3).
• The Algorithm on pages 12 and 13 must be contained on one page and not span between two pages.
• Table 12 must not span between two pages.
• The authors must include a separate Discussion section before the Conclusion section.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
1. The abstract is long and not focused, you need to rewrite it, make it shorter and concise.
2. The motivation is weak and needs more elaboration in the introduction section
3. I would like to see a clear link between the listed contributions and the sections of the paper.
4. Add a summary table at the end of Section 2 to list all features of the similar work in the literature compared to the proposed one.
5. There are multiple mobility graph related publications in the literature, such as https://ieeexplore.ieee.org/abstract/document/9714731 so, how did the authors filter out these work from the literature review of this work? What are the criteria they used to select the related literature studies?
6. Figure 1 is too small to read.
7. You also need to add a table of notations used in the paper
8. Figure 4 is poor and does not contain any helpful details
9. Please double check the output of Algo1, as it does not match the output put at the top of the algorithm.
10. What is the metric value of the y-axis in Fig9?
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 4 Report
Major Revision
This work presents a method for Remote Sensing Scene Graph and Knowledge Graph Matching with Parallel Walking Algorithm.
The work is an ordinary approach which uses existing software packages and combines them to achieve the wanted results.
The paper is well organized with all required material. i.e.
1. A very extensive review of the relevant literature
2. Comparisons with existing techniques
3. In depth analysis of the way the method can be applied.
4. Conclusions that cover the achieved goals.
However, there are significant problems:
1. The numbers of the cited references in the text are given in a non-formal and syntactically incorrect way. Do not start a sentence with [2] or [16].
2. The Figure caption should be more descriptive and in some way self-standing.
3. Please take care of punctuation almost all over the text, as well as in the captions.
4. Please adapt references as required by Remote Sensing guidelines.
5. Please add two or three recent (2022) references.
6. Tables 3 to 6 can be smaller so that they appear in an ordinary format in the text. Please shorten columns width.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 5 Report
The paper is interesting and the manuscript is well-written. Results support the conclusions and the authors' claims. Descriptions are exhaustive, and globally the proposal is sound. The provided experiments are enough to validate the initial hypothesis.
So, in my opinion, the paper may be accepted.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Dear Authors,
The authors took into account all the reviewers suggestions and observations, and made the necessary corrections.
Some minor suggestions have been made in the article
eg
It is recommended to check that the title of the tables is in accordance with the Microsoft Word template, Remote Sensing journal
Styles: MDPI_4.1_table_caption
References
“Author 1, A.B.; Author 2, C.D. Title of the article. Abbreviated Journal Name Year, Volume, page range.”
"Remote Sens. 2021, 13, 1312" instead of "Remote Sens. 2021, 13, 1312"
It is recommended to check the entire References chapter and correct it, if necessary.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
The authors covered my comments.
Author Response
Please see the attachment.
Reviewer 3 Report
I have no other comments about this paper. The authors have addressed the previously raised ones. I would advise the authors to highlight the changes they have made in the revised version in future to make it easier for the reviewers to spot these changes.
Author Response
Please see the attachment.
Reviewer 4 Report
The authors complied with my suggestion. Please adapt the fonts in the references according to the Journal requirements (i.e. bold in the year).
Author Response
Please see the attachment.
Author Response File: Author Response.docx