Maize (Zea mays L.) Stem Target Region Extraction and Stem Diameter Measurement Based on an Internal Gradient Algorithm in Field Conditions

Round 1

Reviewer 1 Report

Your paper is well written, has a good structure that is easy to follow and is clear in what it describes. The topic you are dealing with is very important for crop breeding development and food security. The stem contour extraction of maize is challenging to measure effectively and efficiently in open field conditions. Your contribution seems to be promising  given your accuracy results. With respect to the remote sensing related components of your study, your methodology is clear and straightforward, without anything surprising. 

I did have some questions regarding the replicability of your study in different settings (different growing stages and different varieties of the plant) but you addressed these by yourselves in the discussion section when referring to limitations. You also stated the limitation of monitoring only the target region and diameter of maize stems, without obtaining growth information from other parts of the plant. 

I have a few more comments:

Line 60: You could say a few things about the Time of Flight camera if that is what you mean here. Use Time of Flight (ToF) when you first refers to this.

Line 66: Here you can also say a few words about this this type of camera. Something basic and simple such as  that it provides depth and color as output at the same time.

Line 231: Mean Absolute Error, Mean Absolute Percentage Error (MAPE), Root Mean Square  Error (RMSE), and Coefficient of Determination (R²) are broadly used. I do not believe that you need to provide background for those. 

The R of the Coefficient of Determination should be written in italics.

General comment: Check the quality of your figures. For example Figure 8 looks like it can be improved. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper is well-written and technically sound. The only revision I suggest is to enhance the quality of the figures. 

(1) Regarding Figure 2, the caption has to be modified.

(2) For Figure 8, the legends are difficult to recognize.

(3) I suggest the font in the figures be the same as the one in the main body.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

In the present work, the authors proposed a phenotyping monitoring technology based on the internal gradient algorithm to acquire the target region and the diameter of maize stems. It is known that the target region and the diameter of the maize stems are important phenotypic parameters to evaluate the vitality of the culture and estimate the biomass of the culture. The evaluation results indicated that the internal gradient algorithm is a promising method.

The research topic is relevant and the authors clearly justify the importance of the study carried out. One of the main arguments is that traditional plant phenotyping monitoring technology relies on ruler-based measurements that are inefficient, laborious and subjective. The text emphasizes the importance of proposing efficient and low-cost phenotyping technologies. Indeed, this is a global demand and research in this area is of fundamental importance. The study goes in this direction by adding important information on the subject.

As for the materials used, experimental aspects and statistical methods, I believe that they were efficiently detailed and that they are adequate to the research objectives and the scope of the journal. Limitations regarding the methods employed and even with regard to the variety of material were well pointed out by the authors, and should be addressed in future work. Obviously, considering other analysis methods, other biometric measurements and conducting studies with different genotypes would be fundamental, adding information to make the results and conclusions more solid. However, this does not detract from the merit of this work and raises relevant topics for discussion and implementation possibilities.

The results section was generally well conducted. Figures and tables have been merged into the text in sufficient numbers to allow a good illustration of the main findings. However, some of these figures are not of good quality and this can make interpretation and reading difficult. In addition, there is a loose figure in the text, which clashes with the structure well conducted by the authors. The last table needs a better narrative for the results and the last figure should not be placed directly below the table without text separating the two. It's just that very little has been said about the possible results to be interpreted from the Figure 9 and, even in an aesthetic sense, it leaves something to be desired.

In the “Discussion” section, the authors discuss their findings well, but lacked references, discussing their results with those of other authors. In my opinion, it is necessary for authors to delimit results that already exist in the literature from results that were found in their research. This should be very clear to readers.

In the conclusions, the main issue was duly addressed and the main results were highlighted. However, the authors begin the text adding information that has already been placed in the introduction, objectives and methodology. I believe conclusions should be listed directly and concisely.

The references cited in the text are relevant and represent current knowledge on the topic addressed. The authors use these references well, especially in the introduction that brings credibility to their objectives, justifications and methodology. However, as previously highlighted, in the discussion it would be necessary to emphasize the main contributions of the present study, delimiting what already exists from what was found.

In general, the work makes important contributions and has several positive points and fits the scope of the journal. But for it to be suitable for publication some improvements would be necessary. In this sense, I highlight below some recommendations to improve the quality of the text:

1) In subsection “2.6.2. Image Quality Evaluation Metrics” the authors showed how MSE and PSNR are interpreted (with respect to their values). But the same was not done with SSIM. I suggest putting this information so that it is clear to any reader.

2) Figures 8 and 9 do not have good quality, font size and subtitles are not adequate and the resolutions hinder the interpretations;

3) In subsection “3.3. Error Analysis of Stem Diameter Measurement” it is necessary to better describe the results of Table 1. Also, placing Figure 9 after the table without any text does not look good. I suggest discussing the table, putting a text calling the Figure. I believe that the results illustrated in Figure 9 could be better described.

4) In discussion it is necessary to emphasize the main results in comparison with those of other authors adding references for this. It is necessary to emphasize what has already been done about what was discovered by the present study.

5) In the conclusions, many aspects that were already contemplated in the introduction, objectives and methodology are listed. Conclusions should be stated more concisely and directly.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf