Tomato Disease Classification and Identification Method Based on Multimodal Fusion Deep Learning

Round 1

Reviewer 1 Report

Dear,

I read the manuscript “agriculture-1993839”, the work has quality, but still needs to be restructured in some points that are below.

Minor revisions

a. Introduction (Lines – 33-37). Insert some citation in this paragraph. Which country? Insert the name of the Country and the places where you have the largest tomato plantations. Which disease? Please clarify this part of the disease better in this first paragraph, perhaps a citation of some work will help you in this task. As I said, quotes are missing to support what you are saying.

b. Line 41 Lucas et al. [5]? Provide a context for this work before citing it.

c. Line 57-60. You could insert that paragraph in your justification, the last paragraph of the introduction.

d. Complementing the justification of your study, I believe that the last paragraph of the introduction (lines 99-107) cannot be just technical, but explain the real reasons and applications for society as a whole about the method being used/proposed. See that the idea you are bringing to this manuscript is and application, I am not able to verify this application for society, please improve this part.

e. Line 111. What is RGB? Is the spectrum visible? Red; Green and Blue? Please make this clearer in your study. Figure 1 shows that the spectrum is visible.

f. Figure 1. Insert a, b, c, d etc...Create a caption for this and increase the horizontal width of the image. See that your caption information is very impoverished. It should be improved and contain more information.

g. Figure 2. Improve the quality of the Figure and increase the size. I couldn't understand the final flowchart. Would you be able to enter more information? Again the caption should contain more information. I suggest checking the possibility of inserting numerical or alphabetical information in each figure, but first check this in the figure formatting guide on the journal page.

Major Reviews

a. Item 2.1.1. I don't understand how the images were acquired. Sensor type? Spectral range? Please even enter the machine model.

b. Figure 4 is important, but contains very little information about the model. Your caption should contain more information. Please see this possibility to improve your figure and legend at this point.

c. Equation 1 seems to me that the font is different from the text, in addition, the line that should contain the information of the terms of the equations is missing important information. What probability is used? What is each term of this equation? I believe that the authors should do more to improve this physical and mathematical information that is being used in this work.

d. Ditto for equation 2;

e. I couldn't understand Figure 6. What is the temperature and humidity information used? What are the green boxes? Please note that it is a scientific work and this information must be accompanied and well explained in this work. What unit is used for temperature? Is it air or leaf temperature? Ditto for the Humidity variable? Is it leaf or air humidity? Improve the Figure caption as well.

f. I didn't understand lines 277-278? Is it your computer? Is this information necessary? Wouldn't it be more appropriate to describe the camera as a spectral scan? See this in all your work.

g. Equation 3 has the same problem as the other equations. Note that the formatting does not follow the same pattern as the text. Improve the information of the terms of the equation.

h. Did you develop the Kappa coefficient? Is there any font missing here? See that some equations are missing this information!!! Check better the description of this coefficient and its involvement in this work, it is not clear enough.

i. Figure 8 should be re-edited. I believe there is important information in it, but the sources are extremely small and crude in their quality. I suggest using the R software and transforming some figures into Boxplot analysis. Subtitles should be improved and well explained.

j. What is the scientific contribution to society? Leave this passage at the conclusion.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This article is about the classification and identification of Tomato disease based on multimodal fusion deep learning. Really the article is complete in every way. I have several suggestions:

1. Please insert the time of training models in the text.

2. Please explain about LabelMe software.

3. Please explain the ratio train, validation, and test sets from all data.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors,

Thanks for the responses and revisions to your paper, so I believe the bibliographic citation of the Kappa index used in your paper is still missing, please provide that citation. I accept the explanation of Figure 8, but see that there are more specific software that can improve the quality of your Figure, there are still some details that I believe are important before disclosing your work to the international community, therefore, I suggest that you insert the slashes sides of the X and Y axes of this Figure and who see the possibility of placing all fonts the same, the text (a and b) seems different to me from the Figures created.

Kind regards,

Author Response

Please see the attachment.

Author Response File: Author Response.pdf