Evaluating the Canopy Chlorophyll Density of Maize at the Whole Growth Stage Based on Multi-Scale UAV Image Feature Fusion and Machine Learning Methods

Round 1

Reviewer 1 Report

Dear authors, thank you for submitting your work to Agriculture. Please, in order to move forward provide answers/corrections to the following points:

1.- Must follow proper template directions for references through the text.

2.- Could it be possible to highlight the novelty of this work? Overall, this kind of research has been widely reported.

3.- You stated that yield and quality are key and related to food security, from there you linked this statement with nitrogen, then with chlorophyll content. However, no yield or quality was reported during this study. Why these two main variables linked to chlorophyll according to your literature review were not studied?

4.- You mentioned in the conclusions that this research can provide technical and data support for precision agriculture. Explain exactly which technical support is derived from your work. Provide concrete examples.

5.- while RF was the best algorithm, still not clear if this is related to the yield and quality, therefore, while computationally interesting, no novelty on real applications was found in your work. Include further directions in this research line.

Author Response

We would like to thank the reviewer for their constructive and helpful suggestions and improvements to our manuscript (Manuscript ID: agriculture-2295548). We revised the manuscript by following the suggestions of the reviewer.The detailed response has been listed in the submitted document (Response to Reviewer1.docx).

Author Response File: Author Response.docx

Reviewer 2 Report

The work needs some adjustments to be approved.

1. The theme of the work is very relevant, but there is a need to improve the article in the scientific context.

2. The TITLE of the paper suggests a study to estimate the chlorophyll density of the maize canopy at all growth stages using multiscale UAV imaging resource fusion and machine learning methods. A suggestion would be to put the word EVALUATION or ANALYSIS in the Title.

3. The article's ABSTRACT must clearly present the objective of the work. The objective of the article is not to MONITOR but to EVALUATE OR ANALYZE.

4. The INTRODUCTION is well written, but I noticed that the purpose of the article needs to come before the last paragraph.

5. In MATERIALS AND METHODS, I see that a methodology flowchart would be very positive for the article.

6. THE RESULTS are very positive.

7. THE DISCUSSION OF THE RESULTS presents interesting statements. See if you can insert a few more bibliographic citations to corroborate the results of the study.

8. CONCLUSIONS are good.

Author Response

We would like to thank the reviewer for their constructive and helpful suggestions and improvements to our manuscript (Manuscript ID: agriculture-2295548). We revised the manuscript by following the suggestions of the reviewer.The detailed response has been listed in the submitted document (Response to Reviewer2.docx).

Author Response File: Author Response.docx

Reviewer 3 Report

Estimating Canopy Chlorophyll Density of Maize at the Whole Growth Stage Using Multi-scale UAV Image Feature Fusion and Machine Learning Methods

The topic of the work is in general interestnig and the paper well communicated.

Nevertheless a revision is needed before publication.

Here below the main issues: 

1) in the experimental part, an eneven condition is evident with many points that are giving evidence of low/very low CCD: such biased condition might have produced an apparently higher performance of the methods. the authors should discuss this points and in partcular the presence of an uneven dataset. 

2) The work is carried out in a specific area with quite small plots: in such conditions, characterized by a relatively low variability, it is quite "easy" to achieve good performance from models. Issues arise whenever such model are extended to different areas or different weather conditions. authors should discuss on the way the model might be used, explaining how they could be implemented in a corn field under different management or environmental conditions.

3) regrding the 14 varieties, the athors should better describe the differences and the different results highlighted by different varieties. 

4) the work makes use of a relevant amount of data. It would be useful to discuss how much processing/computational time has been needed for post processing for the different approaches (see e.g.: Rahman et al, doi: 10.3390/agriculture9010017) and how many data they have been using (as a measure of the dgitization footprint, see e.g.: Kayad et al.: doi: 10.1016/j.compag.2022.107080)

5) plots are very small and might exihibit border effects: did you notice border effects? can such small plots be representative of the actual behaaviour in open field conditions? please comment in the paper. 

6) in the verification phase, some points are clearly deviating from the model. Authors should try to explain with moree details why such eveents are occurring: are they outliers, or how can they be explained? please report with more detail in the paper.  

7) was fertilization and irrigation homogeneous over the field? please provide more specifications

8) please provide information about ground resolution and scanned area

9) how much time took the total scanning of the area? 

10) was incident light corrected? or how was spectral data calibrated?

Author Response

We would like to thank the reviewer for their constructive and helpful suggestions and improvements to our manuscript (Manuscript ID: agriculture-2295548). We revised the manuscript by following the suggestions of the reviewer.The detailed response has been listed in the submitted document (Response to Reviewer3.docx).

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear authors, thank you for reviewing my comments. I know I went hard with you, but I am very happy you were able to handle the comments in the best possible way.

Regards.

Reviewer 3 Report

The paper has been improved in agreement with the comments of the referee, and is now ready for publication