A Distance Measurement Approach for Large Fruit Picking with Single Camera

Round 1

Reviewer 1 Report

The paper "A distance measurement approach for large fruit picking with single camera" presents a research where a method for monocular vision was tested on fruits of multiple ranges in order to be used by a robot which will pick them automatically. The results triggered an absolute error of less than 0.06 m. The outcomes will be valuable for further scientific researches.

- Originality/Novelty: The work is original, as a new method of object multi-range detection was presented. The proposed solution can be further used in research. 

- Significance: The results of the research are interpreted properly. The conclusions are justified and support by the results which are displayed in tables, as well as graphically, along with descriptions and comparisions, but they need to be enriched, because they occupy a lot of space, but they are not well analyzed. All hypotheses were specified and the outcomes need to be carefully analyzed and compared. 

- Quality of Presentation: The article is written appropriately, respecting the logical succession of sections. Data and analyses are presented graphically and inside tables.

The authors may include a new figure at the end of the introduction which illustrates the proposed approach graphically.

The authors need to further emphasize the limitations of existing works in order to prove the originality of their contribution.

Avoid to have paragraphs that occupy half of the page, as in section 1.

The Introduction section of the paper needs to mention the structure of the current article.

Do not add dots after figure names.

The name of the title needs to appear below the image, not on a different page, as it is the case for Figure 4.

Please perform a comparison of the results from Table 1. Do not provide a phrase like = "Which model had a more exact predicting ability than the other one would be justified by counting up the number of every grade.". This sounds like the read should figure it out by himself.

Section 3 starts on line 179, at the end of page 5, but the content is placed on the next page. Please place everything together on page 6.

All four images of Figure 6 need to appear together. Not on two pages.

Page 7 cannot contain just one figure and a table. No explanations are present on the page.

Table 2 needs to appear on one page. Do not split it on two pages. Same applied for Table 3 which is splitted on three pages and at the beginning of every page, the table header is unknown.

The explanations for equation 10 need to appear on the same page, not on a different one.

Place Figure 10 on one page. Please explain it.

Mention the software and the hardware details of the study. On what kind of computer were the tests performed? Which are its characteristics?

The references section must be expanded and include more recent papers from the last 3 years.

- Scientific Soundness: The study of the large fruit picking research approach is well documented and will be useful for further researches.

- Interest to the Readers: The conclusions will surely interest the readers of the Horticulturae Journal, and not only them, as robots with various tasks become adopted by more and more domains.  

- Overall Merit: The benefit to publish this paper consists in a good documented research regarding a fruit picking robot. 

- English Level: The level of English language is advanced. Through the entire paper, the language was appropriate and understandable, being easy to follow the flow since the beginning.

Author Response

Dear Reviewer,

Thanks a lot for your suggestions. We have read and accepted your suggestions about our work and corrected the errors in article. The new version of article has been uploaded. The following is the reply and explanation for that. Please let us know if any revision needed, we would like to do our best to make the manuscript better.

Sincerely,

Authors

The paper "A distance measurement approach for large fruit picking with single camera" presents a research where a method for monocular vision was tested on fruits of multiple ranges in order to be used by a robot which will pick them automatically. The results triggered an absolute error of less than 0.06 m. The outcomes will be valuable for further scientific researches.

- Originality/Novelty: The work is original, as a new method of object multi-range detection was presented. The proposed solution can be further used in research.

- Significance: The results of the research are interpreted properly. The conclusions are justified and support by the results which are displayed in tables, as well as graphically, along with descriptions and comparisons, but they need to be enriched, because they occupy a lot of space, but they are not well analyzed. All hypotheses were specified and the outcomes need to be carefully analyzed and compared.

- Quality of Presentation: The article is written appropriately, respecting the logical succession of sections. Data and analyses are presented graphically and inside tables.

The authors may include a new figure at the end of the introduction which illustrates the proposed approach graphically.

REPLY: A flowchart figure has been added in the end of Introduction to describe the process of distance prediction.

The authors need to further emphasize the limitations of existing works in order to prove the originality of their contribution.

REPLY: In the new versions of this article, the last 3 paragraphs of Introduction, which were added in red, has explained this question.

Avoid to have paragraphs that occupy half of the page, as in section 1.

REPLY: Has been corrected.

The Introduction section of the paper needs to mention the structure of the current article.

REPLY: there were a paragraph added from Line 87 to 94 that “In this work, the principle of monocular distance measurement was first studied and the used material and methods was introduced in Section 2; and then in Section 3.1, the results of establishing models using square checkerboards and circular checkerboards of different areas was shown, followingly, the fruits pictures of 2 oblate pomelos, 2 pyriform pomelos and 2 oval watermelons were made to verify the accuracy of the 2 established models; Via error analysis, the 2 established models were modified to reduce the error, the results were shown in Section 3.2; Finally the whole work and the future expectations was concluded Section 4. The flowchart of this work was shown as Figure 1.”

Do not add dots after figure names.

REPLY: Has been corrected.

The name of the title needs to appear below the image, not on a different page, as it is the case for Figure 4.

REPLY: Has been corrected.

Please perform a comparison of the results from Table 1. Do not provide a phrase like = "Which model had a more exact predicting ability than the other one would be justified by counting up the number of every grade.". This sounds like the read should figure it out by himself.

REPLY: Has been corrected.

Section 3 starts on line 179, at the end of page 5, but the content is placed on the next page. Please place everything together on page 6.

REPLY: Has been corrected.

All four images of Figure 6 need to appear together. Not on two pages.

REPLY: Has been corrected.

Page 7 cannot contain just one figure and a table. No explanations are present on the page.

REPLY: Has been corrected.

Table 2 needs to appear on one page. Do not split it on two pages. Same applied for Table 3 which is split on three pages and at the beginning of every page, the table header is unknown.

REPLY: Has been corrected.

The explanations for equation 10 need to appear on the same page, not on a different one.

REPLY: Has been corrected.

Place Figure 10 on one page. Please explain it.

REPLY: Has been corrected.

Mention the software and the hardware details of the study. On what kind of computer were the tests performed? Which are its characteristics?

REPLY: the relative information was added to Line 168: “The style of used computer was LAPTOP-5V7TD45J,with a CPU of AMD Ryzen 5 3500U with Radeon Vega Mobile Gfx(2.10 GHz).” This was just a kind of common and cheap computer, but it could be fully satisfied for computing requirement.

The references section must be expanded and include more recent papers from the last 3 years.

REPLY: Has been added 4 new references from the last 3 years.

- Scientific Soundness: The study of the large fruit picking research approach is well documented and will be useful for further researches.

- Interest to the Readers: The conclusions will surely interest the readers of the Horticulturae Journal, and not only them, as robots with various tasks become adopted by more and more domains. 

- Overall Merit: The benefit to publish this paper consists in a good documented research regarding a fruit picking robot.

- English Level: The level of English language is advanced. Through the entire paper, the language was appropriate and understandable, being easy to follow the flow since the beginning.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present a method for measuring the distance of large fruits based on monocular vision, which can be useful for picking tasks. The method uses images of checkerboards to establish models that relate the number of pixels occupied by the target area and the imaging distance. The models are tested with images of different shapes of watermelons and pomelos.

1. Please, make it clear what is new in your work compare with previous works; something else that:  line 301 -> "The results indicated that the proposed method provided a new thought to achieve accurate distance measurement for fruit picking device."

2. Lines 16, 17, 18, 126, 127, etc. -> Separate the unity from number.

3. Line 16: "cm2" -> use superscripts -> cm²

4. Line 22: What is "frronit"? Front? Please, use a spelling checker.

5. Line 39: "(MIT) ," there is a space between the parenthesis and the comma.

6. Line 42, 43 and others quotations: "Naoki Yamaguti et al." -> use the magazine's style rules -> [15] or "Yamaguti et al. (1997) /// et al. in italics.

7. Line 128 and others: "Figure. 2" -> delete the period -> "Figure 2".

8. In Tables and Figures include units.

9. You can summarize the Figure 6 in two graphs.

10. Rewrite the section Conclusions. Please, do not repeat the results; draw conclusion from the results.

11. In some parts of your paper, you write about the background where the fruits are but, in your work, you do not include the background as a factor.  Clarify this, please.

12. Line 110 and others -> remove the indentation.

Author Response

Dear Reviewer,

Thanks a lot for your suggestions. We have read and accepted your suggestions about our work and corrected the errors in article. The new version of article has been uploaded. The following is the reply and explanation for that. Please let us know if any revision needed, we would like to do our best to make the manuscript better.

Sincerely,

Authors

Comments and Suggestions for Authors from Reviewer 2

1. Please, make it clear what is new in your work compare with previous works; something else that: line 301 -> "The results indicated that the proposed method provided a new thought to achieve accurate distance measurement for fruit picking device."

REPLY: In the new versions of this article, the last 3 paragraphs of Introduction, which were added in red, has explained this question.

2. Lines 16, 17, 18, 126, 127, etc. -> Separate the unity from number.

REPLY: Has been corrected.

3. Line 16: "cm2" -> use superscripts -> cm2

REPLY: Has been corrected.

4. Line 22: What is "frronit"? Front? Please, use a spelling checker.

REPLY: Has been corrected.

5. Line 39: "(MIT) ," there is a space between the parenthesis and the comma.

REPLY: Has been corrected.

6. Line 42, 43 and others quotations: "Naoki Yamaguti et al." -> use the magazine's style rules -> [15] or "Yamaguti et al. (1997) /// et al. in italics.

REPLY: Has been corrected.

7. Line 128 and others: "Figure. 2" -> delete the period -> "Figure 2".

REPLY: Has been corrected.

8. In Tables and Figures include units.

REPLY: Has been added.

9. You can summarize the Figure 6 in two graphs.

REPLY: Has deleted two graphs due to they performed the similar meaning with the remaining two.

10. Rewrite the section Conclusions. Please, do not repeat the results; draw conclusion from the results.

REPLY: The Conclusions were polished again and marked as red words.

11. In some parts of your paper, you write about the background where the fruits are but, in your work, you do not include the background as a factor. Clarify this, please.

REPLY: First, the environmental background in images could be got rid of according to the previous work we have finished; Besides, due to the specialty of growing fruits, we thought the monocular vision as the most adoptable method for measurement detection. And it could well be combined with Visual Servoing System.

12. Line 110 and others -> remove the indentation.

REPLY: Has been corrected.

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript was written in a manner that conforms to standard scientific method. A reasonable number of references are cited as evidence of substantive background. The topic covered in the article is interesting both from a scientific and a technical point of view, as it provides new and promising knowledge, which can improve, from a technical and economic point of view, the development of automated systems for fruit harvesting. The methods and experimental procedures are described with enough detail to enable future researchers to follow-up aspects of the authors’ work.  The results are presented in a clear manner using graphical and table means. Therefore, I believe that the paper can be published in the journal Horticolturae in the presented form

Author Response

Dear Reviewer,

Thanks for your approvement and approving for our work. We will absorb your valuable suggestions to polish our following work. The new version of article has been uploaded.

Sincerely,

Authors

Author Response File: Author Response.docx

Reviewer 4 Report

In the manuscript, the authors report a method for ranging large size fruit based on monocular vision, which is cost-effective and less-computation involving, as the authors claimed. Two distance prediction models are proposed, based on the images of the square (S)checkerboards and circular (C) checkerboards. The accuracy of the models S and C are tested by using large size watermelon and pomelo fruits of varying shapes. The group further showed that the circular checkerboard model is more accurate than the square checkerboard model.

From my point of view, the topic of this work is very good and is highly relevant to horticulture. The scientific content of the manuscript and the presentation of the results in the figures and tables is also very good. But the overall scientific writing including English writing of the manuscript must be improved significantly. In addition, I would recommend that the authors give satisfactory response to the concerns mentioned below before further consideration. 

Major concerns:

1. Section '1. Introduction'. In this section, the authors fail to highlight the importance of their work. It would be beneficial for the readers if the authors add a separate paragraph highlighting the novelty and importance of their work.

2. Page 4. Line 148. What does the authors mean by the phrase "Their number of pixels and distance were recorded analogously."

3. A flowchart must be added to clearly describe the process of distance prediction.

Minor comments:

1. Page 1. Line 16. In the line "... with 100cm2, 121cm2, 144cm2, 169cm2...), it should cm² (square centimeter) instead of cm2.

2. Page 1. Line 22. What is meant by 'frronit'?

3. Page 4. Lines 156, 157. The sentence "The prediction results of distance could be got if the number of pixels and the area are substituted to the corresponding models respectively." is difficult to understand.

4. Page 14. Lines 290, 297. It is sufficient to mention the equation numbers, instead of writing the whole equation.

Author Response

Dear Reviewer,

Thanks a lot for your suggestions. We have read and accepted your suggestions about our work, and corrected the errors in article. The new version of article has been uploaded. The following is the reply and explanation for that. Please let us know if any revision needed, we would like to do our best to make the manuscript better.

Sincerely,

Authors

In the manuscript, the authors report a method for ranging large size fruit based on monocular vision, which is cost-effective and less-computation involving, as the authors claimed. Two distance prediction models are proposed, based on the images of the square (S)checkerboards and circular (C) checkerboards. The accuracy of the models S and C are tested by using large size watermelon and pomelo fruits of varying shapes. The group further showed that the circular checkerboard model is more accurate than the square checkerboard model.

From my point of view, the topic of this work is very good and is highly relevant to horticulture. The scientific content of the manuscript and the presentation of the results in the figures and tables is also very good. But the overall scientific writing including English writing of the manuscript must be improved significantly. In addition, I would recommend that the authors give satisfactory response to the concerns mentioned below before further consideration.

Major concerns:

1. Section '1. Introduction'. In this section, the authors fail to highlight the importance of their work. It would be beneficial for the readers if the authors add a separate paragraph highlighting the novelty and importance of their work.

Reply: There a separate paragraph added in Introduction paragraph to highlight the novelty and importance of their work.:

‘Although the application of monocular vision in fruit and vegetable picking was rare, its imaging system was simple, lightweight, small in size, did not require precise image matching, and would have advantages in fast computing speed, which not only enabled the cost of subsequent machine vision servo systems to be controlled within an acceptable price, but also facilitated the development of multi-robot collaborative working machines and could provide a reference for the identification of target depth information in the pro-duction process of other fruits and vegetables or crops.’

2. Page 4. Line 148. What does the authors mean by the phrase "Their number of pixels and distance were recorded analogously."

Reply: Has been corrected as "Their number of pixels and distance were recorded automatically."

3. A flowchart must be added to clearly describe the process of distance prediction.

Reply: A flowchart has been added in the end of Introduction to describe the process of distance prediction.

Minor comments:

1. Page 1. Line 16. In the line "... with 100cm2, 121cm2, 144cm2, 169cm2...), it should cm2 (square centimeter) instead of cm2.

REPLY: Has been corrected.

2. Page 1. Line 22. What is meant by 'frronit'?

REPLY: Has been corrected.

3. Page 4. Lines 156, 157. The sentence "The prediction results of distance could be got if the number of pixels and the area are substituted to the corresponding models respectively." is difficult to understand.

REPLY: Has been corrected to “The predicted distance could be got when the number of pixels and the projected area of fruits were input to the models”

4. Page 14. Lines 290, 297. It is sufficient to mention the equation numbers, instead of writing the whole equation.

REPLY: Has been corrected.

Author Response File: Author Response.docx

Round 2

Reviewer 4 Report

I am satisfied with the responses given by the authors and corresponding corrections in the manuscript. I recommend the acceptance of the manuscript.