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.
Thanks for your interest in the journal.
[# PeerJ Staff Note - this decision was reviewed and approved by Xiangjie Kong, a PeerJ Section Editor covering this Section #]
all of my concerns have been addressed.
no comment
no comment
All my concerns have been addressed. I recommend the paper for publication.
All my concerns have been addressed. I recommend the paper for publication.
All my concerns have been addressed. I recommend the paper for publication.
All my concerns have been addressed. I recommend the paper for publication.
Several reviewers have now commented. Overall, you should be sure to address the comments of Reviewers 3 and 4 in detail
[# 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 #]
Manuscript is well written using scientific English language. Objectives and methodology are clearly defined. Background and literature survey is adequate. References used are from relevant indexed journals and conferences. Script is supported by neatly drawn figures and result tables.
It indicates original research.
(1) The C2f module in the YOLOv8n backbone is replaced by C2f-deformable convolutional network (DCN) module to raise the detection accuracy of irregular fish targets.
(2) The Dual-pooling squeeze-and-excitation (DPSE) attention mechanism is put forward and integrated into the YOLOv8n neck network to reinforce the features of the visible parts of the occluded fish target.
(3) Small detection is introduced to make the network more capable of sensing small targets and improving recall.
(4) To improve the bounding box regression performance Wise intersection over union (IOU) rather than the original loss function is used.
Experimentation is done by individual addition of attention mechanisms, loss function and different detection algorithms.
One major contribution by authors is YOLOv8n-DDSW target detection algorithm to resolve the detection difficulties resulting from fish occlusion, deformation and detail loss in complex intensive aquaculture scenarios.
All underlying data have been provided; they are robust, statistically sound, & controlled.
Results, Discussions and conclusions are well stated.
Overall it’s a good work and contributes to the field of automation in underwater object detection.
Well written and easy to follow generally.
Literature review/introduction is too short, need to be enhanced with more references on YOLOv8 modification and applications analyses.
Suitable experiments were conducted.
It would be more interesting to see application in another similar datasets in fisheries for results robustness evaluation of the proposed algorithm.
There should be additional metrics to see the real-time performance of the model such as flops/parameter count and fps assessment. this can be added in Table 4 or new table.
The manuscript proposes YOLOv8n-DDSW, an improved version of the YOLOv8 network tailored for fish target detection in dense underwater environments. The authors introduce several modifications, including deformable convolutions, a dual-pooling squeeze-and-excitation (DPSE) attention mechanism, a small target detection head, and a new Wise-IOU loss function. The work addresses an practical issue in aquaculture monitoring, but several aspects of the study need significant improvement.
1. Dataset Scale and Selection. The dataset used in this study contains only 629 images, which is insufficient to assess the model's robustness and generalizability. Additionally, the dataset focuses on a single species in a single environment. Consider incorporating additional public datasets related to fish detection, such as DeepFish, or expanding the dataset to include multiple species and environments to make the study more comprehensive.
2. Dataset Split Issues. The current dataset is divided into training, validation, and test sets in a 7:2:1 ratio, leaving only ~60 images for testing. This is insufficient to evaluate the model’s generalization in real-world scenarios. I suggest adopting a spatially stratified sampling strategy or ensuring a larger test set for meaningful evaluation.
3. Limited Baseline Comparisons The comparison is limited to YOLOv8 and its variants. The manuscript would be significantly strengthened by including two-stage object detection methods like Faster R-CNN and SSD, as well as other state-of-the-art one-stage detectors.
4. Resource Efficiency Evaluation. While the authors highlight improvements in accuracy, there is no discussion of computational efficiency. Considering the additional modules introduced (e.g., deformable convolutions and DPSE), the paper should report the parameter count, inference time, training time, and GPU memory usage for all compared methods.
no comment
1. Motivation for Choosing YOLO-Based Models. The authors justify the choice of YOLO by emphasizing speed. However, they do not explicitly state whether real-time performance is a critical requirement for the task. If not, explain why YOLO was preferred over two-stage methods, which generally offer better accuracy.
2. Justification for Additional Modules. The manuscript introduces several modules, such as deformable convolutions and DPSE, without providing sufficient motivation. Why were these specific modules selected? I recommend visualizing feature maps before and after applying these modules to show their contribution to improved detection performance.
3. Literature Review. The review of related work is incomplete. Recent advances in underwater detection and remote sensing, such as those using Mamba architectures or specific methods for underwater imagery enhancement, are not discussed, including [0.1109/TGRS.2025.3526927], [10.1109/TGRS.2024.3417253], and [10.1109/DTPI61353.2024.10778881].
[# 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 #]
Since underwater imagery can also be considered remote sensing, it would be beneficial to include insights from general remote sensing literature.
This paper introduces a efficient fish target detection network for dense underwater scenes. Many experiments are used to verify the effectiveness of the proposed method. The following points could be considered.
The motivation and novelty in the abstract and Introduction should be improved.
In section 2, the C2f-DCN module needs more detailed theoretical analysis and mathematical derivation. More ablation studies are needed to justify the specific design choices of the DPSE attention mechanism.
The dataset size (629 images) is insufficient for deep learning model training. More details about data augmentation techniques are needed.
The comparison with state-of-the-art methods is limited. More recent fish detection algorithms in comparisons are needed. Comparison of computational costs and inference times are suggested to be included.
Analysis of model performance under different underwater conditions is missing. Performance on varying fish sizes and orientations should be analyzed. Impact of water quality on detection performance is not analyzed.
More qualitative results showing detection performance are needed.
Current limitations and real-world application constraints are not adequately discussed.
Training protocol needs more detailed description. Inference optimization strategies should be discussed. More visual results are needed.
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.