Estimation of Agronomic Characters of Wheat Based on Variable Selection and Machine Learning Algorithms
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors obtained images using a UAV containing an RGB and hyperspectral camera. Afterwards, AGB and LNC values of wheat were estimated with different algorithms and machine learning methods. The study is successful in terms of application. Here are some of my suggestions for the article.
1) Explanations of AGB and LNC abbreviations should be added to the abstract.
2) The information about previous studies in the introduction section is insufficient.
3) Information about the importance and definition of wheat is required for the introduction section. Some studies that can be used for this:
https://doi.org/10.3390/rs15143595
https://doi.org/10.1007/s12161-022-02251-0
https://doi.org/10.3390/rs14205141
https://doi.org/10.1007/s00217-022-04029-4
4) The processing steps in the image processing section should be shown on a sample image.
5) On page 8, the same paragraph is written twice.
6) What the representation in Figure 4 means should first be explained. The reader should recognize this image.
7) What are the advantages and disadvantages of using satellite images (for example, Sentinel 2) instead of UAV? Discuss.
8) What is the reason why deep learning methods are not applied? Discuss.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Authors,
I have gladly accepted the role of reviewing the paper on the potential use of hyperspectral and RGB data for predicting wheat yield and nitrogen content in various growth stages. Here are a few comments on the text:
It seems worthwhile to revise the introduction and include some values observed during data analysis. The introduction doesn't indicate which indicators were considered or the perspective from which they were analyzed.
In line 151, the authors use the term "hyperspectral lens," which I don't entirely understand. Please consider changing this misleading term. A similar issue was noted in line 157, where the authors use the incorrect term "Lens calibration." I kindly request a thorough review by the authors before resubmitting the manuscript for review.
It would be much easier to read statistical values if they were presented in a table rather than on individual figures.
I've noticed several shortcomings in the text that need to be addressed before the article is published:
Throughout sections 2.3 and 2.4, there is no mention of the spatial resolution of hyperspectral data. Additionally, there is no presentation of the calibration method for RGB data with hyperspectral data. In Figure 2, there is an observed misalignment between RGB and hyperspectral data, and the method and data for regression are not provided.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsComments on the previous article version were responded to appropriately by the authors. However, Comment 7 and Comment 8 should be discussed within the article.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsI have gladly received the text after the authors made revisions. It seems to me that the most significant errors have been corrected, which has significantly improved the manuscript.
However, I propose making a few additional corrections:
Figure 2 requires improvement and the addition of information regarding radiometric and geometric calibration.
I would enhance Figure 5 by removing excessive information from the individual plots. In this case, the plots contain too much data, which hampers their readability.
In the text, there are imprecise statements, such as in rows 168-169, "Lens calibration was required to adjust the exposure time before the start of the mission." The authors likely meant radiometric camera calibration before launch. Therefore, I suggest reviewing the text for such linguistic inaccuracies.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
