Next Article in Journal
Effect of High Temperatures on the Growth and Disease Development of Erysiphe quercicola on Rubber Trees
Previous Article in Journal
Application of Compost as an Organic Amendment for Enhancing Soil Quality and Sweet Basil (Ocimum basilicum L.) Growth: Agronomic and Ecotoxicological Evaluation
 
 
Article
Peer-Review Record

Apple Pest and Disease Detection Network with Partial Multi-Scale Feature Extraction and Efficient Hierarchical Feature Fusion

Agronomy 2025, 15(5), 1043; https://doi.org/10.3390/agronomy15051043
by Weihao Bao and Fuquan Zhang *
Reviewer 1: Anonymous
Reviewer 2:
Reviewer 3:
Agronomy 2025, 15(5), 1043; https://doi.org/10.3390/agronomy15051043
Submission received: 18 March 2025 / Revised: 17 April 2025 / Accepted: 24 April 2025 / Published: 26 April 2025
(This article belongs to the Section Pest and Disease Management)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper is interesting and well written. It represents a further step forward to the use of YOLO in the detection of pests and diseases. The text is detailed and does not neglect any aspect by comparing the results achieved with those of other works and models. To further improve the article, the authors should improve the quality of the presentation by adjusting some figures and captions. I have indicated in the specific comments the changes to be made.

This paper needs minor revision.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The article is interesting and the analyses have been carried out correctly. I have included my specific comments in the text.

The authors should pay more attention to adhering to the editorial requirements of the editors.

 

In particular, the authors should try to indicate the correct journal in which they wish to publish their work. The reviewer received the article from the journal “Agronomy” while the manuscript descriptions indicate the journal “Forest” from the same publisher. The reviewer asked the publisher to explain in depth whether there was an attempt at publishing fraud here.

For this reason, I reject the article in its current form.

I have included substantive and editorial comments in the text of the article.

Due to the large number of parameters, variables, their designations and units of measurement included in the text, I suggest creating a notation table at the beginning of the paper containing an alphabetical list of them (according to the journal's editorial guidelines).

They should also pay more attention to the titles given to the subsections of the paper.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

 

The use of machine learning in current research on pests and diseases affecting various plants is widespread. The algorithms applied in studies on the appearance of diseased plant tissues are increasingly effective at recognizing damage and responding appropriately and swiftly to its visual signs. This paper addresses a very important issue concerning the detection of apple leaf diseases. It seems that further research is necessary to improve both the quality and speed of disease detection.

Here are a few comments on the manuscript:

Regarding the overall concept of the work, I have a few minor remarks about the choice of the dataset. The authors refer to only four diseases, which are presented in the dataset. In my opinion, the authors should also seek out other datasets with varying parameters, such as resolution, and compare the model’s performance across these datasets.

In the title of the paper, the authors suggest that the work focuses solely on leaf diseases, whereas the text also discusses diseases manifesting on fruits and plant shoots. It may be worth adjusting the title to better reflect this scope.

I suggest refining Figure 13 to better illustrate the model's recognition accuracy compared to YOLOv11. The figure indicates that in some cases, the YOLOv11 model shows better recognition of infected areas than the proposed YOLO-PEL model.

Comments on the Quality of English Language

In some places, the text is unclear and overly complex. I recommend simplifying the descriptions of each data processing stage.

Author Response

请参阅附件。

Author Response File: Author Response.docx

Back to TopTop