Impact of Leaf Occlusions on Yield Assessment by Computer Vision in Commercial Vineyards

Round 1

Reviewer 1 Report

The work deals with an interesting and current problem, already extensively explored also by some of the authors of the work.
The methodology used is correct but the results are fairly well known.
The future possibility of using multiband sensors (NIR, IR) could be mentioned.

References from 24 onwards are missing

In the text, there is an excessive (inappropriate) use of some terms (eg: commercial) that could be eliminated.

Comments for author File: Comments.pdf

Author Response

After reading the reviewer’s comments, we have carefully modified and improved the work based on their suggestions. We first explain the general changes and then we provide a more detailed explanation to the reviewer’s comments.

The main changes introduced in our new manuscript are the following:

• All new text content (modifications, new sentences, data, etc.) was highlighted in blue in the document to facilitate its tracking.

• All text was carefully reviewed and improved for better readability.

• References have been revised and modified.

• Some general comments of Reviewer 1

Reviewer_1 (RW-1). The work deals with an interesting and current problem, already extensively explored also by some of the authors of the work.
The methodology used is correct but the results are fairly well known.
The future possibility of using multiband sensors (NIR, IR) could be mentioned.

Authors: We appreciate your comments. Applications of multiband sensors has been mentioned in the Discussion. Some references have been added.

RW-1. In the text, there is an excessive (inappropriate) use of some terms (eg: commercial) that could be eliminated.

Authors: We thank you for this suggestion. The term “commercial” has been cancelled in many sentences in the manuscript, it has been used too often in the text. However, we would prefer to keep this term in the title to indicate that our developments and technologies can be applied by grape and wine industry. We think it is a new tool for grapegrowers.

• Specific comments of Reviewer 1

RW-1. References from 24 onwards are missing.

Authors: References have been revised and modified.

Author Response File: Author Response.docx

Reviewer 2 Report

Interesting study on a topic that needs to be addressed. In my opinion, more work needs to be completed in commercial settings like this. It would be interesting to see other canopy configurations tested in a similar way. 

Lines 38-44 need some references. Line 38 is vague, 39 lists several other applications but none are referenced. 

Line 59 mentions cool climate viticulture, but the region of interest is not. Warm region viticulturists often use leaf removal for similar applications. There are plenty of studies on this issue in warm climates. 

Line 84: Avoid starting sentences with numbers. What does it mean vines were chosen in "any commercial vineyard"? Does this mean you randomly chose vineyards or that there were 25 vines in "any" of your target vineyards. Rephrase.

Line 152: what are "very good results"? I get that this is a methods section, not results, but this sentence is vague. 

Line 186: I appreciate that this work was done in a commercial setting where the issue of vine to vine variability is a real issue. 

Figures probably need a consistent colour scheme for text and axes. 

Unless I am missing it somewhere, there is no code for significance in the figures. 

Line 291-292: The use of "no defoliated vineyards" reads as if you didn't do the machine vision in defoliated vineyards. I would suggest using either an acronym system (ND, PD, FD) with your treatments or to reword (Leaf occlusion rate was determined by machine vision in vineyards with no canopy defoliation). 

Line 339-341: This sentence is odd and very speculative. You want growers to remove leaves if they want yield prediction? I would consider this very restrictive. Just because there are benefits to leaf removal doesn't mean that every does it (or do it willingly), though in this region it may be near 100% (in others it most certainly is not 100%) Economics will drive this. If growers pay for leaf removal it will certainly make it easier for yield prediction, but would you ask a grower that wasn't going to remove leaves to do it just so they can get a yield estimate? If so, how much benefit do you see to the grower in knowing the final yield, especially this close to harvest? Some of the literature reviewed looked at berries much earlier in the season for this reason. There is usually less economic value to the grower knowing the final yield the longer the season goes on. Suggest reworking this section. 

Overall, I think you have a nice study here and should be ready for publication after addressing referee comments.

Author Response

After reading the reviewers’ comments, we have carefully modified and improved the work based on their suggestions. We first explain the general changes and then we provide a more detailed explanation to the reviewers’ comments.

The main changes introduced in our new manuscript are the following:

• All new text content (modifications, new sentences, data, etc.) was highlighted in green in the document to facilitate its tracking.
• Figures has been modified.

• All text was carefully reviewed and improved for better readability.

Specific comments of Reviewer 2

Reviewer_2 (RW-2) Interesting study on a topic that needs to be addressed. In my opinion, more work needs to be completed in commercial settings like this. It would be interesting to see other canopy configurations tested in a similar way.

Authors: We thank you for these comments. We agree to test other canopy configurations and grapevine varieties in the future. It should be interesting.

RW-2. Figures probably need a consistent colour scheme for text and axes.

Authors: We thank you for this annotation. Figures have been modified based on the reviewers’ suggestions.

RW-2. Unless I am missing it somewhere, there is no code for significance in the figures.

Authors: The significance has been included in the figures

RW-2. Overall, I think you have a nice study here and should be ready for publication after addressing referee comments.

Authors: We appreciate these comments. Thank you!

RW-2. Lines 38-44 need some references. Line 38 is vague, 39 lists several other applications but none are referenced.

Authors: Some references have been included.

RW-2. Line 59 mentions cool climate viticulture, but the region of interest is not. Warm region viticulturists often use leaf removal for similar applications. There are plenty of studies on this issue in warm climates.

Authors: We agree to these comments. The text was modified.

RW-2. Line 84: Avoid starting sentences with numbers. What does it mean vines were chosen in "any commercial vineyard"? Does this mean you randomly chose vineyards or that there were 25 vines in "any" of your target vineyards. Rephrase.

Authors: The text has been modified.

RW-2: Line 152: what are "very good results"? I get that this is a methods section, not results, but this sentence is vague.

Authors: The text has been modified

RW-2. Line 186: I appreciate that this work was done in a commercial setting where the issue of vine to vine variability is a real issue.

Authors: We thank you for this comment. A large variability was observed in commercial vineyards.

RW-2. Line 291-292: The use of "no defoliated vineyards" reads as if you didn't do the machine vision in defoliated vineyards. I would suggest using either an acronym system (ND, PD, FD) with your treatments or to reword (Leaf occlusion rate was determined by machine vision in vineyards with no canopy defoliation).

Authors: We have carefully modified and improved the text based on these suggestions

RW-2. Line 339-341: This sentence is odd and very speculative. You want growers to remove leaves if they want yield prediction? I would consider this very restrictive. Just because there are benefits to leaf removal doesn't mean that every does it (or do it willingly), though in this region it may be near 100% (in others it most certainly is not 100%) Economics will drive this. If growers pay for leaf removal it will certainly make it easier for yield prediction, but would you ask a grower that wasn't going to remove leaves to do it just so they can get a yield estimate? If so, how much benefit do you see to the grower in knowing the final yield, especially this close to harvest? Some of the literature reviewed looked at berries much earlier in the season for this reason. There is usually less economic value to the grower knowing the final yield the longer the season goes on. Suggest reworking this section.

Authors: We have carefully modified and improved the text based on these suggestions

Author Response File: Author Response.docx