Dear Dr. Abbod,

Please note that, at this level of publication, we have see that, one or to reviewers have asked for particular citations, this is not mandatory for the authors, it is the responsability of the authors to check if the forced references are relevant to the study or not. the authors are not forced to report any citation that is not relevant to the study and a detail reply with solid justification should be provided.

Please submit your revised manuscript by Aug 21 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Salim Heddam

Academic Editor

PLOS ONE

Journal requirements: 

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf .

2. Please note that PLOS ONE has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, we expect all author-generated code to be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse.

3. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

The American Journal Experts (AJE) (https://www.aje.com/) is one such service that has extensive experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. Please note that having the manuscript copyedited by AJE or any other editing services does not guarantee selection for peer review or acceptance for publication.

Upon resubmission, please provide the following:

The name of the colleague or the details of the professional service that edited your manuscript

A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

A clean copy of the edited manuscript (uploaded as the new *manuscript* file)”.

4. We note that your Data Availability Statement is currently as follows: [All relevant data are within the manuscript and its Supporting Information files.]

Please confirm at this time whether or not your submission contains all raw data required to replicate the results of your study. Authors must share the “minimal data set” for their submission. PLOS defines the minimal data set to consist of the data required to replicate all study findings reported in the article, as well as related metadata and methods (https://journals.plos.org/plosone/s/data-availability#loc-minimal-data-set-definition).

For example, authors should submit the following data:

- The values behind the means, standard deviations and other measures reported;

- The values used to build graphs;

- The points extracted from images for analysis.

Authors do not need to submit their entire data set if only a portion of the data was used in the reported study.

If your submission does not contain these data, please either upload them as Supporting Information files or deposit them to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of recommended repositories, please see https://journals.plos.org/plosone/s/recommended-repositories.

If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent. If data are owned by a third party, please indicate how others may request data access.

5. When completing the data availability statement of the submission form, you indicated that you will make your data available on acceptance. We strongly recommend all authors decide on a data sharing plan before acceptance, as the process can be lengthy and hold up publication timelines. Please note that, though access restrictions are acceptable now, your entire data will need to be made freely accessible if your manuscript is accepted for publication. This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If you are unable to adhere to our open data policy, please kindly revise your statement to explain your reasoning and we will seek the editor's input on an exemption. Please be assured that, once you have provided your new statement, the assessment of your exemption will not hold up the peer review process.

6.  Please include a copy of Table 5 which you refer to in your text on page 23.

Comments from the Editorial Office:

One or more of the reviewers has recommended that you cite specific previously published works. Members of the editorial team have determined that the works referenced are not directly related to the submitted manuscript. As such, please note that it is not necessary or expected to cite the works requested by the reviewer. 

Furthermore, please do not respond to Reviewer 1 as we have concerns that have utilised AI tools in their review.

Reviewer 1#:

This manuscript presents an innovative approach for optimizing hyperparameters of the YOLOv10 deep learning model for arson detection. The authors propose a hybrid optimization algorithm that combines the Grey Wolf Optimization (GWO) and Brown Bear Optimization Algorithm (BBOA), referred to as GWO-BBOA. This hybrid algorithm aims to improve the model's precision and recall in detecting arson events by fine-tuning hyperparameters such as learning rate, momentum, and weight decay. The method was evaluated on a custom arson detection dataset of 2,182 images. The results indicate that GWO-BBOA outperforms traditional optimization methods such as PSO, GWO, and BBOA, showing better performance in precision, recall, and mAP (mean Average Precision).

Significance of the Study:

The study highlights the growing importance of real-time anomaly detection systems, particularly for safety-critical applications such as arson detection. By optimizing YOLOv10 using a hybrid optimization approach, the manuscript demonstrates the potential of advanced machine learning and optimization techniques in improving model performance for real-time applications. The contribution is significant, as it addresses the challenge of balancing high precision and low false-positive rates in a domain where false alarms can have serious consequences.

Content and Structure Review

Abstract:

The abstract effectively summarizes the key contributions of the study. It clearly outlines the methodology and provides quantitative results (precision, recall, and mAP), which is essential for understanding the model's performance. However, it could be more concise in stating the results and their implications. Furthermore, it would be helpful to briefly mention the limitations of the current model, such as false positives and scalability issues, which are addressed in the manuscript.

Introduction:

The introduction provides a strong rationale for the study, clearly stating the importance of accurate arson detection in high-risk areas. However, the manuscript would benefit from a more detailed comparison with existing methods and an explanation of why YOLOv10 was selected as the model of choice over other object detection algorithms. Additionally, references to recent works on hybrid optimization algorithms, such as DOI: 10.54216/MOR.030205, would strengthen the argument for the effectiveness of the hybrid approach.

Conclusion:

The conclusion summarizes the key findings well. However, it could provide more insight into future work. For example, the authors could discuss the potential for integrating this model with real-time fire detection systems, expanding the dataset, or improving the model's performance under varying environmental conditions.

Literature Review and Citation Updates

Literature Review:

The literature review provides a comprehensive background on arson detection and optimization algorithms used in deep learning models. It discusses the limitations of previous studies and justifies the need for a hybrid optimization approach. However, it would benefit from the inclusion of more recent works on hybrid algorithms in deep learning, such as DOI: 10.54216/JAIM.090102, which could provide further context to the hybrid methodology employed in this study.

Citations:

The manuscript includes a solid list of references, but it could benefit from the inclusion of more recent studies on hybrid machine learning approaches, such as DOI: 10.54216/MOR.030205, to provide a broader perspective on the evolution of optimization techniques in deep learning models.

https://doi.org/10.1016/j.eswa.2023.122147

https://doi.org/10.54216/JAIM.090102

https://doi.org/10.54216/MOR.030205

https://doi.org/10.1007/s11540-024-09717-0

https://doi.org/10.32604/cmc.2023.031723

Technical Review

Methodology and Algorithms:

The methodology is well-described, with a clear explanation of the proposed hybrid optimization algorithm, GWO-BBOA. However, more detailed explanations of the individual algorithms (GWO, BBOA, and their hybridization) would improve the clarity of the methodology. For example, a more thorough explanation of the mechanism behind the pedal scent marking and sniffing behaviors in BBOA would help readers understand why these steps are important in the optimization process.

Hyperparameter Tuning and Validation:

The manuscript briefly mentions hyperparameter tuning but lacks sufficient detail about the exact search space used for each parameter. Including more details about how the hyperparameters were tuned and how the results were validated (e.g., through cross-validation) would improve the rigor of the study. The authors could also discuss the computational cost of training with these optimization techniques.

Performance Evaluation

Result Presentation:

The results section provides a clear presentation of the performance metrics for the proposed model. However, additional performance metrics such as confusion matrices and ROC curves would help assess the model's classification performance in more detail. Visual comparisons of the proposed GWO-BBOA method with other algorithms would also enhance the reader’s understanding of the model's strengths and weaknesses.

Visualizations:

The manuscript includes some useful tables and figures, but additional visualizations comparing the results of different algorithms would improve the presentation. A graphical representation of precision-recall curves or a heatmap of mAP scores would be particularly helpful.

Reviewer 2#:

The paper presents a promising hybrid optimization approach for arson detection using YOLOv10. However, it requires substantial revision to improve clarity, methodological transparency, and critical discussion of limitations and applicability. Addressing these comments will significantly enhance the quality and impact of the manuscript.

Overall Strengths:

The paper tackles an important real-world problem (arson detection) using recent advancements in deep learning and optimization.

It is well-structured and mostly adheres to academic conventions.

The hybrid GWO-BBOA algorithm is a novel contribution with demonstrated performance gains.

Extensive experiments and comparisons are made, including model variants and multiple optimization strategies.

Comments for Improvement:

Clarity and English Language Quality

The manuscript contains grammatical and syntactic issues throughout. Many sentences are long, redundant, or awkwardly phrased. A professional English language editing service is highly recommended.

Example: “Besides, damage early minimizes loss of lives, obedience to authorities…” → This is unclear. Likely intended to be something like: “Early damage detection minimizes loss of life and ensures compliance with safety regulations.”

Suggestion - Please send for proofread.

Methodological Justification

The choice of hyperparameters (e.g., lr0, lrf, mo, wd) should be better justified based on domain knowledge or ablation studies.

The rationale behind specific values (e.g., population size = 20, epochs = 20) should be clarified—were these empirically determined?

Novelty and Comparison

While the hybrid GWO-BBOA shows better results than individual algorithms, it’s unclear how statistically significant these differences are. A statistical significance test (e.g., t-test or ANOVA) should be included.

The comparison with Singh et al. (2020) is weak, as the models, datasets, and objectives differ greatly. Instead, a comparison with similar object detection or fire/arson detection papers using YOLO variants would strengthen the discussion.

Lack of Visual Results

There is a notable absence of qualitative visual results (e.g., bounding box detection outputs for arson scenes). Including detection outputs for different scenarios (e.g., low light, occlusions) would demonstrate practical robustness.

Dataset Limitations and Augmentation

Although augmentation expands the dataset, the original number of 290 frames from 53 videos is still quite small for training deep models. This limitation needs to be acknowledged and discussed more thoroughly, especially its effect on model generalization.

Real-World Deployment Discussion

The conclusion briefly mentions integration with real-time systems. However, more elaboration is needed: What are the latency constraints? Can the model run on edge devices (e.g., security cameras)? What challenges are foreseen in such deployments?

Minor Comments and Technical Corrections:

Abstract: Avoid listing too many metrics in the abstract unless absolutely necessary.

Introduction: The statement "Accidental fires can also cause disastrous effects such causing fires on purpose" is unclear and contradictory.

Section 2.3.4: The pseudocode is helpful but could benefit from line numbers and a figure/table format for clarity.

References: Make sure the reference style is consistent. Some citations are numbered inline; others are formatted inconsistently.

Reviewer 3#:

While results are convincing, cross-validation or testing on a secondary dataset would enhance the generalizability of the claims.

Including confidence intervals or statistical significance tests could further strengthen the validity of the comparisons.

Single-run evaluation is implied (no mention of repeated trials or random seeds), which can limit the reproducibility and reliability of the reported performance.

Reviewer 4#:

Strengths:

1. The paper addresses an important and practical problem: arson detection in real-time surveillance settings using deep learning techniques.

2. The proposed hybrid optimization method (GWO-BBOA) applied to YOLOv10 is interesting, and experimental results are reported to support its effectiveness.

Major Concerns and Suggestions for Improvement:

1. Lack of Related Work Section:

The manuscript does not include a dedicated Related Work section. A detailed comparison with prior work—especially in the areas of fire/arson detection and other optimization methods—is crucial for positioning the novelty and contribution of this work.

2. Insufficient Discussion of Novelty and Challenges:

The introduction briefly states the use of YOLOv10 and the hybrid optimization method, but it does not clearly articulate the technical challenges being addressed that prior works have not solved. Please elaborate on:

a) What specific limitations in previous models or optimization strategies this work aims to overcome?

b) Why GWO-BBOA is particularly suitable in this context?

3. No Visualization of Detection Results:

To assess the practical effectiveness of the proposed method, the paper should include visual examples of detection results (e.g., bounding boxes overlaid on frames with fire/arson events). This helps reviewers and readers evaluate qualitative performance.

4. Lack of Model Input/Output Description:

The paper does not clearly explain what kind of data is used as input to the model (e.g., image resolution, frame rate, RGB vs. thermal, etc.) and what format the model outputs. Please provide a clearer overview of the pipeline, including preprocessing (if any) and output interpretation.

5. No Code Availability:

To ensure reproducibility and to support future research, the authors should provide a public link to their code. This is particularly important for works involving novel optimization strategies applied to established models.

Reviewer 5#:

The manuscript explores hyperparameter optimization strategies to enhance the performance of YOLOv10, which is a lightweight yet fast real-time object detection model. The authors proposed a new method, GWO-BBOA, that integrates Grey Wolf Optimization (GWO) and the Brown Bear Optimization Algorithm (BBOA) to effectively tune the model's hyperparameters.

Comments:

The reproduction of Figure 2 from the original YOLOv10 paper is not presented clearly. All four figures are in low resolution, which affects readability. It is strongly recommended to use high-resolution or vector graphics when applicable.

Baselines are essential for evaluation. Please include the training results of YOLOv10 using the original hyperparameters in Table 3 for comparison.

It seems that two pictures augmented from the same original picture were split across the training set and test sets. Arson002_x264_frame_0060.jpg from the test set and Arson002_x264_frame_0059_aug_out_aug_1.png from the training set are very similar. To prevent data leakage and ensure the validity of results, such overlap should be avoided.

In line 339, it’s unclear whether 100 epochs were applied to all training. Since different hyperparameters may lead to the model convergence at different epochs. It’s advisable to report the stopping strategy used in this research. Reporting metrics such as mAP or loss at each epoch would support the experimental results.

Regarding Table 4, more relevant and recent works may serve as better points of comparison. Suggested alternatives include: https://doi.org/10.1016/j.eswa.2023.119741 and https://doi.org/10.3390/app14135841.

The program mentioned on line 138 needs to be properly cited.

Grammatical correctness is needed at lines 35, 149, and 333.

A range should be given at each line of 389, 390, and 392.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: N/A

**********

3. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

**********

Reviewer #1: This manuscript presents an innovative approach for optimizing hyperparameters of the YOLOv10 deep learning model for arson detection. The authors propose a hybrid optimization algorithm that combines the Grey Wolf Optimization (GWO) and Brown Bear Optimization Algorithm (BBOA), referred to as GWO-BBOA. This hybrid algorithm aims to improve the model's precision and recall in detecting arson events by fine-tuning hyperparameters such as learning rate, momentum, and weight decay. The method was evaluated on a custom arson detection dataset of 2,182 images. The results indicate that GWO-BBOA outperforms traditional optimization methods such as PSO, GWO, and BBOA, showing better performance in precision, recall, and mAP (mean Average Precision).

Significance of the Study:

The study highlights the growing importance of real-time anomaly detection systems, particularly for safety-critical applications such as arson detection. By optimizing YOLOv10 using a hybrid optimization approach, the manuscript demonstrates the potential of advanced machine learning and optimization techniques in improving model performance for real-time applications. The contribution is significant, as it addresses the challenge of balancing high precision and low false-positive rates in a domain where false alarms can have serious consequences.

Content and Structure Review

Abstract:

The abstract effectively summarizes the key contributions of the study. It clearly outlines the methodology and provides quantitative results (precision, recall, and mAP), which is essential for understanding the model's performance. However, it could be more concise in stating the results and their implications. Furthermore, it would be helpful to briefly mention the limitations of the current model, such as false positives and scalability issues, which are addressed in the manuscript.

Introduction:

The introduction provides a strong rationale for the study, clearly stating the importance of accurate arson detection in high-risk areas. However, the manuscript would benefit from a more detailed comparison with existing methods and an explanation of why YOLOv10 was selected as the model of choice over other object detection algorithms. Additionally, references to recent works on hybrid optimization algorithms, such as DOI: 10.54216/MOR.030205, would strengthen the argument for the effectiveness of the hybrid approach.

Conclusion:

The conclusion summarizes the key findings well. However, it could provide more insight into future work. For example, the authors could discuss the potential for integrating this model with real-time fire detection systems, expanding the dataset, or improving the model's performance under varying environmental conditions.

Literature Review and Citation Updates

Literature Review:

The literature review provides a comprehensive background on arson detection and optimization algorithms used in deep learning models. It discusses the limitations of previous studies and justifies the need for a hybrid optimization approach. However, it would benefit from the inclusion of more recent works on hybrid algorithms in deep learning, such as DOI: 10.54216/JAIM.090102, which could provide further context to the hybrid methodology employed in this study.

Citations:

The manuscript includes a solid list of references, but it could benefit from the inclusion of more recent studies on hybrid machine learning approaches, such as DOI: 10.54216/MOR.030205, to provide a broader perspective on the evolution of optimization techniques in deep learning models.

https://doi.org/10.1016/j.eswa.2023.122147

https://doi.org/10.54216/JAIM.090102

https://doi.org/10.54216/MOR.030205

https://doi.org/10.1007/s11540-024-09717-0

https://doi.org/10.32604/cmc.2023.031723

Technical Review

Methodology and Algorithms:

The methodology is well-described, with a clear explanation of the proposed hybrid optimization algorithm, GWO-BBOA. However, more detailed explanations of the individual algorithms (GWO, BBOA, and their hybridization) would improve the clarity of the methodology. For example, a more thorough explanation of the mechanism behind the pedal scent marking and sniffing behaviors in BBOA would help readers understand why these steps are important in the optimization process.

Hyperparameter Tuning and Validation:

The manuscript briefly mentions hyperparameter tuning but lacks sufficient detail about the exact search space used for each parameter. Including more details about how the hyperparameters were tuned and how the results were validated (e.g., through cross-validation) would improve the rigor of the study. The authors could also discuss the computational cost of training with these optimization techniques.

Performance Evaluation

Result Presentation:

The results section provides a clear presentation of the performance metrics for the proposed model. However, additional performance metrics such as confusion matrices and ROC curves would help assess the model's classification performance in more detail. Visual comparisons of the proposed GWO-BBOA method with other algorithms would also enhance the reader’s understanding of the model's strengths and weaknesses.

Visualizations:

The manuscript includes some useful tables and figures, but additional visualizations comparing the results of different algorithms would improve the presentation. A graphical representation of precision-recall curves or a heatmap of mAP scores would be particularly helpful.

Reviewer #2: The paper presents a promising hybrid optimization approach for arson detection using YOLOv10. However, it requires substantial revision to improve clarity, methodological transparency, and critical discussion of limitations and applicability. Addressing these comments will significantly enhance the quality and impact of the manuscript.

Overall Strengths:

The paper tackles an important real-world problem (arson detection) using recent advancements in deep learning and optimization.

It is well-structured and mostly adheres to academic conventions.

The hybrid GWO-BBOA algorithm is a novel contribution with demonstrated performance gains.

Extensive experiments and comparisons are made, including model variants and multiple optimization strategies.

Comments for Improvement:

Clarity and English Language Quality

The manuscript contains grammatical and syntactic issues throughout. Many sentences are long, redundant, or awkwardly phrased. A professional English language editing service is highly recommended.

Example: “Besides, damage early minimizes loss of lives, obedience to authorities…” → This is unclear. Likely intended to be something like: “Early damage detection minimizes loss of life and ensures compliance with safety regulations.”

Suggestion - Please send for proofread.

Methodological Justification

The choice of hyperparameters (e.g., lr0, lrf, mo, wd) should be better justified based on domain knowledge or ablation studies.

The rationale behind specific values (e.g., population size = 20, epochs = 20) should be clarified—were these empirically determined?

Novelty and Comparison

While the hybrid GWO-BBOA shows better results than individual algorithms, it’s unclear how statistically significant these differences are. A statistical significance test (e.g., t-test or ANOVA) should be included.

The comparison with Singh et al. (2020) is weak, as the models, datasets, and objectives differ greatly. Instead, a comparison with similar object detection or fire/arson detection papers using YOLO variants would strengthen the discussion.

Lack of Visual Results

There is a notable absence of qualitative visual results (e.g., bounding box detection outputs for arson scenes). Including detection outputs for different scenarios (e.g., low light, occlusions) would demonstrate practical robustness.

Dataset Limitations and Augmentation

Although augmentation expands the dataset, the original number of 290 frames from 53 videos is still quite small for training deep models. This limitation needs to be acknowledged and discussed more thoroughly, especially its effect on model generalization.

Real-World Deployment Discussion

The conclusion briefly mentions integration with real-time systems. However, more elaboration is needed: What are the latency constraints? Can the model run on edge devices (e.g., security cameras)? What challenges are foreseen in such deployments?

Minor Comments and Technical Corrections:

Abstract: Avoid listing too many metrics in the abstract unless absolutely necessary.

Introduction: The statement "Accidental fires can also cause disastrous effects such causing fires on purpose" is unclear and contradictory.

Section 2.3.4: The pseudocode is helpful but could benefit from line numbers and a figure/table format for clarity.

References: Make sure the reference style is consistent. Some citations are numbered inline; others are formatted inconsistently.

Reviewer #3: While results are convincing, cross-validation or testing on a secondary dataset would enhance the generalizability of the claims.

Including confidence intervals or statistical significance tests could further strengthen the validity of the comparisons.

Single-run evaluation is implied (no mention of repeated trials or random seeds), which can limit the reproducibility and reliability of the reported performance.

Reviewer #4: Strengths:

1. The paper addresses an important and practical problem: arson detection in real-time surveillance settings using deep learning techniques.

2. The proposed hybrid optimization method (GWO-BBOA) applied to YOLOv10 is interesting, and experimental results are reported to support its effectiveness.

Major Concerns and Suggestions for Improvement:

1. Lack of Related Work Section:

The manuscript does not include a dedicated Related Work section. A detailed comparison with prior work—especially in the areas of fire/arson detection and other optimization methods—is crucial for positioning the novelty and contribution of this work.

2. Insufficient Discussion of Novelty and Challenges:

The introduction briefly states the use of YOLOv10 and the hybrid optimization method, but it does not clearly articulate the technical challenges being addressed that prior works have not solved. Please elaborate on:

a) What specific limitations in previous models or optimization strategies this work aims to overcome?

b) Why GWO-BBOA is particularly suitable in this context?

3. No Visualization of Detection Results:

To assess the practical effectiveness of the proposed method, the paper should include visual examples of detection results (e.g., bounding boxes overlaid on frames with fire/arson events). This helps reviewers and readers evaluate qualitative performance.

4. Lack of Model Input/Output Description:

The paper does not clearly explain what kind of data is used as input to the model (e.g., image resolution, frame rate, RGB vs. thermal, etc.) and what format the model outputs. Please provide a clearer overview of the pipeline, including preprocessing (if any) and output interpretation.

5. No Code Availability:

To ensure reproducibility and to support future research, the authors should provide a public link to their code. This is particularly important for works involving novel optimization strategies applied to established models.

Reviewer #5: The manuscript explores hyperparameter optimization strategies to enhance the performance of YOLOv10, which is a lightweight yet fast real-time object detection model. The authors proposed a new method, GWO-BBOA, that integrates Grey Wolf Optimization (GWO) and the Brown Bear Optimization Algorithm (BBOA) to effectively tune the model's hyperparameters.

Comments:

The reproduction of Figure 2 from the original YOLOv10 paper is not presented clearly. All four figures are in low resolution, which affects readability. It is strongly recommended to use high-resolution or vector graphics when applicable.

Baselines are essential for evaluation. Please include the training results of YOLOv10 using the original hyperparameters in Table 3 for comparison.

It seems that two pictures augmented from the same original picture were split across the training set and test sets. Arson002_x264_frame_0060.jpg from the test set and Arson002_x264_frame_0059_aug_out_aug_1.png from the training set are very similar. To prevent data leakage and ensure the validity of results, such overlap should be avoided.

In line 339, it’s unclear whether 100 epochs were applied to all training. Since different hyperparameters may lead to the model convergence at different epochs. It’s advisable to report the stopping strategy used in this research. Reporting metrics such as mAP or loss at each epoch would support the experimental results.

Regarding Table 4, more relevant and recent works may serve as better points of comparison. Suggested alternatives include: https://doi.org/10.1016/j.eswa.2023.119741 and https://doi.org/10.3390/app14135841.

The program mentioned on line 138 needs to be properly cited.

Grammatical correctness is needed at lines 35, 149, and 333.

A range should be given at each line of 389, 390, and 392.

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes:  V. V. Subrahmanyam

Reviewer #4: No

Reviewer #5: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org . Please note that Supporting Information files do not need this step.

Dear Dr. Abbod,

Please submit your revised manuscript by Oct 13 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Salim Heddam

Academic Editor

PLOS ONE

Journal Requirements:

If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. 

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

Reviewer #5: The authors addressed all of my comments, and the updated version shows improvement. However, some low-level mistakes remain evident, which raises concerns that the authors may have rushed to complete the revision.

All figures are still in low resolution. Some characters in Figures 1 and 5 are unreadable. I recommend that the authors double-check these.

Line 385: “Fig 1. The Flowchart of the Hybrid GWO-BBOA Optimization Process.” This should refer to Figure 5, not Figure 1.

The sentence: “We have added the performance of the YOLOv10 model trained with default hyperparameters (i.e., without any optimization) to Table 4 as the Baseline.” I believe the authors meant Table 5, not Table 4.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #3: All comments have been addressed

Reviewer #4: All comments have been addressed

Reviewer #5: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions??>

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

**********

Reviewer #3: Met the minimum standards of the journal.

Almost addressed the issues raised by the reviewers.

Revision version is much better than the original one.

Reviewer #4: I appreciate authors' feedback. All my concerns are addressed. Therefore, I would vote for acceptance.

Reviewer #5: The authors addressed all of my comments, and the updated version shows improvement. However, some low-level mistakes remain evident, which raises concerns that the authors may have rushed to complete the revision.

All figures are still in low resolution. Some characters in Figures 1 and 5 are unreadable. I recommend that the authors double-check these.

Line 385: “Fig 1. The Flowchart of the Hybrid GWO-BBOA Optimization Process.” This should refer to Figure 5, not Figure 1.

The sentence: “We have added the performance of the YOLOv10 model trained with default hyperparameters (i.e., without any optimization) to Table 4 as the Baseline.” I believe the authors meant Table 5, not Table 4.

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #3: Yes:  Venkata Subrahmanyam Vampugani

Reviewer #4: No

Reviewer #5: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org

Optimizing Hyperparameters of YOLOv10 for Arson Detection Using Advanced Optimization Algorithms

PONE-D-25-07942R2

Dear Dr. Abbod

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager®  and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support .

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Salim Heddam

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #5: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions??>

Reviewer #5: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #5: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #5: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #5: Yes

**********

Reviewer #5: All of my concerns have been addressed, and I appreciate the effort they put into clarifying and improving the manuscript.

**********

what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #5: No

**********