Evaluation of Maize Crop Damage Using UAV-Based RGB and Multispectral Imagery

Round 1

Reviewer 1 Report (Previous Reviewer 4)

The manuscript titled "Evaluation of maize crop damage using UAV-based RGB and multispectral imagery" aims to utilize UAV-based images to estimate the damage areas of maize fields in Poland. The current version of the manuscript shows significant improvement compared to the previous one. However, there are still some existing issues that authors still need to address:

-       While the authors successfully highlight the knowledge gap they aim to address, they should provide more detailed objectives in the paper at the end of the introduction section, rather than just one sentence (L109-L112).

-       Although they mentioned fixing or relocating the "mixture of methodology and result contents," I have observed that some mixed content remains evident (e.g., L199-L203, previously L174-178, or L221-228, previously L196-203). These sections represent typical examples of the aforementioned concern.

Therefore, I believe that the manuscript requires further revisions before it can be considered ready for publication.

Author Response

The manuscript titled "Evaluation of maize crop damage using UAV-based RGB and multispectral imagery" aims to utilize UAV-based images to estimate the damage areas of maize fields in Poland. The current version of the manuscript shows significant improvement compared to the previous one. However, there are still some existing issues that authors still need to address:

-       While the authors successfully highlight the knowledge gap they aim to address, they should provide more detailed objectives in the paper at the end of the introduction section, rather than just one sentence (L109-L112).

Authors: The aim of the study was more detailed and specified at the end of the Introduction.

 

-       Although they mentioned fixing or relocating the "mixture of methodology and result contents," I have observed that some mixed content remains evident (e.g., L199-L203, previously L174-178, or L221-228, previously L196-203). These sections represent typical examples of the aforementioned concern.

Authors: We have moved the parts which describe methods applied from Results to Material and Methods.

 

Therefore, I believe that the manuscript requires further revisions before it can be considered ready for publication.

Authors: Thank you very much for all the comments.

Reviewer 2 Report (Previous Reviewer 3)

the topic is very pratical in term of serving the agriculture sector in the area, I have some minore recomendations for such a vaaluable work, which may the authors take in considerations, as follow:

abstract needs to reorganized :briefly describe the main methods.

 place the study in a broad context preifly and highlight why it is important.  the purpose of the work and its significance  should be define. The current state of the research field should be carefully reviewed .

The Materials and Methods should be described with sufficient details to allow others to replicate and build on the published results. add flowchart of the followed steps.

The result section should provide more interpretation of the results.

discuss the results and how they can be interpreted from the perspective of previous studies and of the working hypotheses .

 

the lanugue is easy to deliver the manuscript message while it contains minor garammer mistakes. review the manuscript.

Author Response

the topic is very pratical in term of serving the agriculture sector in the area, I have some minore recomendations for such a vaaluable work, which may the authors take in considerations, as follow:

abstract needs to reorganized :briefly describe the main methods.

Authors: The abstract was improved to present the methods applied more clearly.

 

 place the study in a broad context preifly and highlight why it is important.  the purpose of the work and its significance  should be define. The current state of the research field should be carefully reviewed .

Authors: We have improved the Introduction to specify more detail the significance of the study.

 

The Materials and Methods should be described with sufficient details to allow others to replicate and build on the published results. add flowchart of the followed steps.

Authors: Material and Methods were improved and better more details are specified. The flowchart was added as Fig. 3.

 

The result section should provide more interpretation of the results.

discuss the results and how they can be interpreted from the perspective of previous studies and of the working hypotheses .

 Authors: Results and Discussion were improved to be more informative.

 

Comments on the Quality of English Language

the lanugue is easy to deliver the manuscript message while it contains minor garammer mistakes. review the manuscript.

 Authors: The text of the manuscript was one more time edited.

Reviewer 3 Report (Previous Reviewer 2)

The methodology used to obtain the ground truth of the areas affected by wild boars remains unclear.

Author Response

The methodology used to obtain the ground truth of the areas affected by wild boars remains unclear.

Authors: The methodology used to obtain the ground truth of the areas affected by wild boars was more detailed described in Material and Methods.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Suggestions for manuscript revision:

1）language revision is needed before the manuscript is ready for publication, although I am not a native English speaker. Statements are hard to understood due to language issue. an examples is listed below: "agriculture-Evaluation of Maize Crop Damage Using UAV-Based RGB and Multispectral Imagery"; 

2)a table with technical parameters of UAV camera will be helpful for understand the data aquasition; 

3)is there any field observed crop damages caused by boar with photos? otherwise we could not be sure that those damages were caused by boar; 

4) "Areas selected by visual assessment of crop damage were treated as the reference for 164

other results obtained in this study.",  this can not ensure the reliability of reference data; 

5) "3.1. Reference areas of crop damage ", should be moved to section 2; 

6) how to distinguish damages caused boar from those caused by other reasons (such as logging) should be discussed, usually there will be more than one cause for crop damage.

1）language revision is needed before the manuscript is ready for publication, although I am not a native English speaker. Statements are hard to understood due to language issue. an examples is listed below: "agriculture-Evaluation of Maize Crop Damage Using UAV-Based RGB and Multispectral Imagery"; 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The methodology used to obtain the ground truth about the affected areas is not very clear. The authors indicate that this was done by visual assessment on orthophotos. Perhaps some graph or more extensive explanation in this regard would be useful.  What criteria did you use to visually identify by photo-interpretation the areas damaged by wild animals from the rest of the damaged areas?

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

This work is significant and of applied importance

Abstract and Conclusions Both are too short and insufficient to express the importance of the work.

fig. 1 should be presented geographically to the country 

Statistical Analysis is insufficient and inexpressive

References are very less especially in the discussion section 

Discussions with previous studies are not sufficient, and it must be noted that creativity in this important work appears through comparison with previous studies

In the end, I wish you good luck

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

 

The manuscript titled "Evaluation of maize crop damage using UAV-based RGB and multispectral imagery" aims to utilize UAV-based images to estimate the damage areas of maize fields in Poland. While the overall structure follows the basic format of a scientific paper, and the study's results have the potential to contribute to the applications of drone imagery in agriculture, there are several current drawbacks that prevent it from being ready for publication in a scientific journal.

Firstly, the introduction fails to effectively highlight the knowledge gap that the research aims to address, despite briefly mentioning the damage caused by wild boars and the use of UAV images in detecting crop damage. Additionally, the study's objective is only presented in a single sentence at the end of the section, which lacks sufficient specificity. A more detailed description of the research objective is needed.

Secondly, the methodology section is too vague and lacks logical coherence. For instance, the authors mention "various classification methods" (L130) but provide no information about them. The next sentence discusses how to select the reference areas. Another example is the "10-fold cross-validation method," (L157) which is mentioned without any reference. While it is a basic technique in machine learning, it is important to include a reference for it to ensure understanding for readers who may be unfamiliar with it.

Thirdly, the results section is disorganized. It includes a mixture of methodology and result contents. For example, the content from L174 to L187, L196 to L203, and L236 to L242 should be placed in the methodology section.

Fourthly, the manuscript leaves many unanswered questions.

- The authors used CART, a machine learning model, for their study, but they haven't mentioned anything about how they split their dataset into training and testing sets to evaluate their machine learning method accurately. If they used the entire dataset for training and testing on the same dataset, it is an incorrect way to apply a machine learning algorithm.

-  Figure 2: Is that the complete dataset for this study, or is it just a subset of it?

- In the CART model, the authors used both NDVI and DSM as inputs. Did the authors try using a single input (NDVI or DSM) and compare it with the threshold they chose? Alternatively, did the authors use CART to find the optimal threshold for NDVI/DSM classification? An alternative strategy for addressing this issue involves utilizing the original RGB images as (additional) inputs for the CART models, which, in my opinion, will enhance the overall accuracy.

Some other issues:

-  L164-166: There is no reference for these statement (Figure 2?)

-  English issues. Example: select or detect? (L236) 

Based on these issues, I regret that I have to reject the paper in its current state.

 

 

Upon reviewing the manuscript, I have identified grammatical and word usage errors. Therefore, I recommend that the authors thoroughly scan the entire manuscript to rectify these issues.

Author Response

Please see the attachment

Author Response File: Author Response.docx