BGM-YOLO: An accurate and efficient detector for detecting plant disease

Chenghai Yu; Junhao Xie; Fernandes Jean Adrian Tony

doi:10.1371/journal.pone.0322750

Peer Review History

Original SubmissionJanuary 24, 2025
24 Jan 2025 Author Response https://doi.org/10.1371/journal.pone.0322750.r001
5 Mar 2025 Decision Letter - Hirenkumar Mewada, Editor Dear Dr. Xie, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Apr 19 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 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, Hirenkumar Kantilal Mewada 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 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. 4. Please ensure that you refer to Figure 2, 4, 5, 6 and 8 in your text as, if accepted, production will need this reference to link the reader to the figure. 5. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 1, 2, 3, and 5 in your text; if accepted, production will need this reference to link the reader to the Table. [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: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: No Reviewer #2: Yes ****** 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 ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ****** Reviewer #1: The manuscript titled “BGM-YOLO: An accurate and efficient detector for detecting plant disease” describes incorporating a incorporates a multiscale bitemporal fusion module to increase the ability of YOLO recognize plant disease. While the authors’ proposed method does appear to increase the accuracy of the model modestly, some improvements need to be made to clarify how the model was compared to previous methods and classes for healthy plants need to be added for each plant species so that it can be shown that the new model can decipher between healthy and sick plants. Major points: 1. The author should specify if pre-trained models were used in the training of their model. 2. The author should show how the accuracy of network increases as more images are provided for training and compare that to the base line models. It would be nice to know how many images are needed for the proposed method to be more accurate than models it is compared to. 3. Each plant species should have a healthy class in addition to diseased classes. If you have diseases that are specific to some plant species in your data, then the model may just be predicting what plant species it is. 4. It is unclear if the model developed in the manuscript has a higher accuracy for identifying the bounding boxes. Something like the Jaccard index should be used to quantify how accurate the bounding boxes are from the models. 5. The results should be split into each plant species with multiple diseases and the healthy classes. This would help prospective users of the model understand how it works across species. 6. Confusion matrices of the test data predictions should be added so the readers understand the nature of the errors the model makes. Minor points: 1. Do table 1 & 2 really need to be tables? You could just state them in materials and methods. 2. Figure 3 is too small to see what is going on in each image. 3. Could the new model better quantify the severity of the disease? 4. Could a well-known pretrained model such as resnet50 to be added to part of the classifier of the new model? Reviewer #2: In the manuscript "BGM-YOLO: An accurate and efficient detector for detecting plant disease", the authors proposed a method to detect plant disease using a deep learning model called BGM-YOLO, addressing challenges such as complex backgrounds, variable lighting, and small-scale disease spots in agricultural environments, etc. The model is based on the YOLOv8n architecture and has 3 key innovations as claimed by the authors: 1) a new GSC2f module which enhances the original C2f module, combining grouped and depthwise separable convolutions to reduce computational costs while maintaining feature extraction efficiency; 2) a multiscale bitemporal fusion module (BFM) which improves feature fusion robustness by leveraging multiscale convolutions and attention mechanisms to balance spatial and channel features across temporal scales dynamically; 3) a median-enhanced spatial and channel attention block (MECS) integrates median pooling with traditional average and max pooling in the channel attention mechanism, coupled with multiscale depthwise convolutions for spatial attention, significantly boosting the detection of small-scale disease features. The model is evaluated on a plant disease object detection dataset from the Roboflow open-source platform which contains over 5k images, and achieves a 4.2% improvement in mean average precision over YOLOv8n, along with reduced false negatives and competitive inference speed. The authors also conducted ablation studies to verify the effectiveness of each module and analyzed the model's performance against several state-of-the-art object detection models. The following are the comments on the manuscript: 1. The numerical experiments used only one dataset, which makes the works less convincing. The authors should consider evaluating the model on more datasets to demonstrate its generalization ability. 2. The authors should provide some statistical information about the dataset, e.g. the classes of the plants and/or diseases this dataset contains (and does it have a highly imbalanced distribution?); the distribution of the bounding boxes (number per image, size, etc.). 3. Some of the figures and their titles are split into two pages. Some of the figures have low resolutions, especially Figure 8. 4. Math equations are not formatted well enough (have significantly larger font size). ****** 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: Yes: Hao WEN ******** [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. https://doi.org/10.1371/journal.pone.0322750.r002
Revision 1
18 Mar 2025 Author Response We sincerely appreciate the valuable feedback and constructive suggestions provided by the reviewers and the editor. These comments have significantly contributed to improving the clarity, accuracy, and completeness of our manuscript. We have carefully addressed all concerns and revised the manuscript accordingly. Below is a summary of our responses to the key points raised: 1.To further validate the generalization ability of our model, we have added Section 3.7, where we conducted additional experiments using an external dataset (Md Faruk Alam’s Tomato Leaf Disease dataset). This dataset differs from our original dataset in terms of disease categories and distribution. To ensure robust evaluation, we expanded the dataset from 700 to 3000 images through data augmentation. The results demonstrate that our model maintains strong generalization capability across different datasets while improving performance in specific categories. 2.As requested, we have explicitly stated in Section 3.2 that no pre-trained models were used in our training process. The model was trained from scratch to learn domain-specific features directly from the plant disease dataset, ensuring better adaptability to the task. 3.In Section 3.6, we analyzed how model accuracy improves as more training images are provided. We conducted experiments using 20%, 40%, 60%, 80%, and 100% of the dataset and evaluated the model's performance. Figure 9 illustrates that the model achieves a mAP50 of X% with 60% of the training data, with diminishing gains beyond this point. This provides insights into the data requirements for optimal performance. 4.We have clarified in Section 3.3 that our evaluation already includes mAP50 and mAP50:95, which inherently assess bounding box accuracy through Intersection over Union (IoU) measurements. These metrics provide a reliable evaluation of the model's object localization performance, as suggested by the reviewer. 5.In Section 3.3, we have expanded the performance analysis to provide class-wise metrics (Precision, Recall, mAP50, mAP50:95) for each plant species and disease category. This allows prospective users to better understand how the model performs across different species and conditions. Table 1 now presents these detailed results. 6.As suggested, we have added Figure 8 in Section 3.3, which presents the confusion matrix for the test dataset. This analysis highlights both correct predictions and misclassifications, revealing potential confusion between visually similar diseases. We also discuss possible solutions, such as enhanced feature extraction techniques, to mitigate classification errors. 7.We have reformatted the manuscript to ensure figures and their titles remain on the same page. Additionally, we have replaced low-resolution images (including Figure 8) with higher-quality versions to enhance clarity and readability. 8.We have adjusted all mathematical equations to ensure consistent font size and proper formatting, improving readability. 9.In response to the reviewer’s suggestion, we have removed Tables 1 & 2 and integrated their content into the text within Section 3.2 for better readability. 10.Our current model focuses on disease detection rather than severity quantification. However, we acknowledge that confidence scores could potentially serve as a proxy for severity assessment. This is noted as a future research direction. 11.We decided not to integrate ResNet50 as an additional classifier to maintain computational efficiency and model simplicity. The YOLO architecture already provides robust classification capabilities, and our experimental results confirm that additional classifiers are unnecessary for achieving high accuracy in this task. We appreciate the reviewers' and editor’s time and effort in evaluating our manuscript. We believe the revisions have significantly improved the paper and hope that it now meets the expectations for publication. Attachments Attachment Submitted filename: Response to Reviewers.docx https://doi.org/10.1371/journal.pone.0322750.r003
28 Mar 2025 Decision Letter - Hirenkumar Mewada, Editor BGM-YOLO: An accurate and efficient detector for detecting plant disease PONE-D-25-03982R1 Dear Dr. Xie, 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. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org. 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, Hirenkumar Kantilal Mewada Academic Editor PLOS ONE Additional Editor Comments (optional): Please refer to the comments provided by the reviewer regarding the revision of your future work section. Their insights could greatly enhance the clarity and impact of your manuscript. I recommend authors revise future work incorporating this suggestion in final draft. Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #1: All comments have been addressed Reviewer #2: All comments have been addressed ******** 2. Is the manuscript technically sound, and do the data support the conclusions??> Reviewer #1: Yes Reviewer #2: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #2: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy Reviewer #1: Yes Reviewer #2: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ****** Reviewer #1: The comments have sufficiently been addressed. It may be worth highlighting that the errors mostly happen within species. Also - spider mites are not a plant disease; they are an insect pest. It would be worth mentioning that the model appears to classify well between insect pest and diseases in that case. Perhaps a future direction of research would be using the model to identify diseases, insect and herbicide damage. These often confuse people who end up sending their plants into plant diagnostic labs. Perhaps a model like yours could tell if damage is from insects, disease or pesticide drift damage in the future. Reviewer #2: In the revised manuscript "BGM-YOLO: An accurate and efficient detector for detecting plant disease", the authors have addressed key concerns raised in the previous reviews, significantly strengthening the methodological rigor and presentation quality. The newly added numerical experiments on the new dataset effectively validates generalization ability and robustness of the proposed model. The manuscript is publication-ready in its current form. ****** 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: Yes: Hao WEN ******** https://doi.org/10.1371/journal.pone.0322750.r004
Formally Accepted
Acceptance Letter - Hirenkumar Mewada, Editor PONE-D-25-03982R1 PLOS ONE Dear Dr. Xie, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. 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. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Hirenkumar Kantilal Mewada Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0322750.r005

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .