All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Dear Author,
Thank you for clearly addressing the reviewer's comments. Your paper seems now sufficiently improved after the second revision. Your manuscript is ready for publication.
Best wishes,
[# PeerJ Staff Note - this decision was reviewed and approved by Xiangjie Kong, a PeerJ Section Editor covering this Section #]
overall, the manuscript quality has been improved despite not all the comments are addressed. e.g. the abstract 'hides' the fact that the authors use old yolov2 as base model.
YOLOv8 has been included for comparison only. imho, the results of the proposed method (using yolov2) does not outperform the one with yolov8.
the results is more like negative results in we benchmark with the later yolo version. perhaps the contribution is more on the semantic aggregation process.
the presentation formatting should be check thoroughly including captions for table and figure, section and subsection title etc
Dear Authors,
Thank you for the reivised paper. One of the previous reviewers accepts your paper as it stands. However, in accordance with the comments provided by the two initial reviewers, it is recommended that the paper undergo a minor revision. It is encouraged that the identified concerns and criticisms be addressed and that the article be resubmitted once the requisite updates have been performed. Reviewer 1 has suggested you could consider specific references. You are welcome to add them if you think they are relevant and useful . However, you are under no obligation to include them, and if you do not, it will not affect my decision.
Best wishes,
[# PeerJ Staff Note: It is PeerJ policy that additional references suggested during the peer-review process should *only* be included if the authors are in agreement that they are relevant and useful #]
The authors proposed a ship detection method that leverages semantic aggregation in complex
backgrounds. Overall the paper is well written and easy to follow. However I have some comments:
- Did the author develop and test custom models beside YOLOv2?
- limitation of the propopsed method needs to be discussed
- Further comparison witth recent state of the art is needed to further motivate the proposed approach. In this regards refer to: An explainable embedded neural system for on-board ship detection from optical satellite imagery; SAR Ship Detection Based on Explainable Evidence Learning under Intra-class Imbalance.
no comment
no comment
no comment
Literature review has been enhanced with additional references.
the authors do not mention sufficiently on why choosing YOLOv2, not mentioning comparison insight with e.g. YOLOv8 0r v9.
no comments
As the YOLO is basically used as the main model, it should be mentioned in the abstract.
in the conclusion, e.g. the proposed semantic aggregation may be implemented in the future for latest YOLO version, and this is as the future recommendation.
The qulity of this paper is increased after revision.
no more comments
no more comments
no more comments
Dear authors,
Thank you for submitting your article. Based on reviews' comments, your article has not yet been recommended for publication in its current form. However, we encourage you to address the concerns and criticisms of the reviewer and to resubmit your article once you have updated it accordingly.
Reviewers have asked you to provide specific references. You are welcome to add them if you think they are relevant. However, you are under no obligation to include them, and if you do not, it will not affect my decision.
Furthermore, equations should be used with correct equation number. Please do not use “as follows”, “given as”, etc. Explanation of the equations should also be checked. All variables should be written in italic as in the equations. Their definitions and boundaries should be defined. Necessary references should be provided. Many of the equations are part of the related sentences. Attention is needed for correct sentence formation.
Best wishes,
[# PeerJ Staff Note: It is PeerJ policy that additional references suggested during the peer-review process should *only* be included if the authors are in agreement that they are relevant and useful #]
no comment'
no comment'
no comment'
This study proposes a ship detection method that leverages semantic aggregation in complex
backgrounds. The paper is interestig. However, I have some comments:
* not clear the innovative contribution of the paper
* limitations of the proposed method need to be discussed
* in relation to the results: did the authors use any validation technique (e.g. k-fold cross validation)? Please, use this technique and refere the results as mean+- standard deviation.
* further comparison with the recent state of the art in the field is needed. Please refer to: An explainable embedded neural system for on-board ship detection from optical satellite imagery; YOLO-OSD: Optimized ship detection and localization in multiresolution SAR satellite images using a hybrid data-model centric approach; High-order spatial interactions enhanced lightweight model for optical remote sensing image-based small ship detection
This study proposes a ship detection method using YOLOv2 and its modification tested on
Seaship dataset. The presentation and analysis are acceptable for journal level.
Some new features contribute to the paper are e.g.:
the metric DIoU, reorg layer (which needs to be elaborated) within the feature fusion module.
However, some areas need to be improved as mentioned in the next comments.
-line 26 (abstract): reorg --> need to replace with better word (easier to understand)
-literature review/introduction --> should include these two works which also use Seaship/ABOship dataset:
https://www.nature.com/articles/s41598-024-64225-y
https://ieeexplore.ieee.org/abstract/document/10576412/
also as literature review: https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/ipr2.12959.
**PeerJ Staff Note:** It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors are in agreement that they are relevant and useful.
Yolov2 is not the latest version of YOLO variants. Authors need to justify of using this older version. while some papers already use yolov5 and even yolov8. see Table 5 of literature survey:
https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/ipr2.12959.
**PeerJ Staff Note:** It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors are in agreement that they are relevant and useful.
-mAP used in this paper needs to be clear, mAP0.5 or other thresholds used.
some comments for improvement:
-benchmark your result with the one in reference above which is using YOLOv8. provide the commentary insight on the results in the conclusion or in discussion. i.e.
https://www.nature.com/articles/s41598-024-64225-y
-many references are older than 5years even 10 years, pls update
**PeerJ Staff Note:** It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors are in agreement that they are relevant and useful.
The paper proposes a ship detection method that leverages semantic aggregation in complex backgrounds. Initially, a semantic aggregation module merges deep features, rich in semantic information, with shallow features abundant in location details extracted via the front-end network. The experimental results demonstrate that the proposed method can enhance the mAP of ship objects while ensuring real-time detection compared to other methods.
It is suggested to give the model size, GFLOPs, FPS of the proposed method in Section Experiments.
1. The motives of this paper's proposed method are unclear.
2. The paper should specify what this fine-grained feature is.
3. The paper contains vague and grammatically incorrect sentences. please carefully correct the errors.
I am pleased to have reviewed this paper. However, I have noticed that the number of references in this paper is relatively low, especially in the Introduction and Related Works sections. I recommend that the authors continue to add a sufficient number of references to demonstrate their extensive research in the field and enhance the credibility of the paper. I understand that finding suitable references can be a challenging and time-consuming task, so I am providing a few references for the authors’ consideration.
1) https://doi.org/10.1016/j.engappai.2023.106271
2) https://doi.org/10.1109/TCSVT.2024.3407057
**PeerJ Staff Note:** It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors are in agreement that they are relevant and useful.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.