Advance of Target Visual Information Acquisition Technology for Fresh Fruit Robotic Harvesting: A Review

Round 1

Reviewer 1 Report

Please improve the work with members, a lot of them is missing.

I have found also some typos, please correct them:

312 binocular camera based  - camera-based

354 contour of stickyand - sticky and

378 optimaloperation - optimal operation

393 It is usually used in robot operation scenarioswith large object distance space - scenarios with

479 effective work space - workspace

513 laser sensor products was - were

 

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled ‘Advance of Target Visual Information Acquisition Technology for Fresh Fruit’s Robotic Harvesting: A Review’ (ID: agronomy-1720546). Those comments (at the bottom of the cover letter) are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval.

The main corrections in the manuscript and the responds to the comments are as following:

Point 1: Please improve the work with members, a lot of them is missing.  

Response 1: The authors have supplemented in the "Author Contributions ". The contents are as follows:

Author Contributions: Conceptualisation, Y.L. and Q.F.; methodology, Y.L. and Q.F.; formal analysis, Q.F., T.L. and C.L.; investigation, Y.L., Q.F. and Z.C.; resources, Y.L. and Q.F.; data curation, Y.L., Q.F. and F.X.; writing—original draft preparation, Y.L. and Q.F.; writing—review and editing, Y.L., T.L., F.X., C.L., and Z.C.; project administration, Q.F.; funding acquisition, Q.F. All authors have read and agreed to the published version of the manuscript.

Point 2: 312 binocular camera based  - camera-based  

Response 2:  Line 312, the statements of ‘binocular camera based’ was corrected as ‘camera-based’.

Point 3: 354 contour of stickyand - sticky and

Response 3:  Line 354, the statements of ‘stickyand’ was corrected as ‘sticky and’. And we have modified the same problem in the full manuscript.

Point 4: 378 optimaloperation - optimal operation

Response 4:  Line 378, the statements of ‘optimaloperation’ was corrected as ‘optimal operation’.

Point 5: 393 It is usually used in robot operation scenarioswith large object distance space - scenarios with

Response 5:  Line 393, the statements of ‘scenarioswith’ was corrected as ‘scenarios with’.

Point 6: 479 effective work space – workspace

Response 6:  Line 479, the statements of ‘work space’ was corrected as ‘workspace’.

Point 7: 513 laser sensor products was – were

Response 7:  Line 513, the statements of ‘was’ was corrected as ‘were’.

 

The above are my main corrections in the paper and the responds to the Reviewer’s comments. Special thanks to you for your comments. We tried our best to improve the manuscript and made some changes in the manuscript. Any revised portion made to the manuscript are marked up using the “Track Changes” function. At the same time, we appreciate for Editors and Reviewers warm work earnestly, and hope that the correction will meet with approval.

 

Once again, thank you very much for your comments and suggestion. If you have any queries, please don’t hesitate to contact me at the address below.

Thank you and best regards.

Yours sincerely,

Qingchun Feng

Author Response File: Author Response.pdf

Reviewer 2 Report

·       The paper has outlined the state-of-the-art vision technology, their current shortcomings, and the potential research trends.

·       The authors have prepared the review by considering the harvesting robots for selected greenhouses and orchards crops. It would have been better if they could review the vision technology used for robotic harvesting of indoor grown crops like mushrooms etc. Then, the paper could be said to cover a good review of vision technology used in robotic harvesting of representative crops grown in all scenarios (indoor and outdoor conditions).

·       The sub-figure 1(f) lacks proper citation.

·       To improve the readability of the article, the various methods used for fruit target identification, location measurement etc and their efficiency could have been presented in a table. And adding the type of hardware used along with these methods will give the reader a better insight. That is what camera sensor (type & configuration) with which fruit identification/localising methods can give a better efficiency for the representative crops mentioned.

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled ‘Advance of Target Visual Information Acquisition Technology for Fresh Fruit’s Robotic Harvesting: A Review’ (ID: agronomy-1720546). Those comments (at the bottom of the cover letter) are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval.

The main corrections in the manuscript and the responds to the comments are as following:

Point 1: The sub-figure 1(f) lacks proper citation.  

Response 1: The authors have supplemented citations to the sub-figure 1(f). The contents are as follows:

Figure 1. Typical harvesting robots. (a) Sweet pepper harvesting robot [10]; (b) Strawberry harvesting robot [16]; (c) Tomato harvesting robot [7]; (d) Apple harvesting robot [15]; (e) Apple harvesting robot [20]; (f) Kiwi harvesting robot [18].

Point 2: To improve the readability of the article, the various methods used for fruit target identification, location measurement etc and their efficiency could have been presented in a table. And adding the type of hardware used along with these methods will give the reader a better insight. That is what camera sensor (type & configuration) with which fruit identification/localising methods can give a better efficiency for the representative crops mentioned.  

Response 2:  

At the beginning of writing the manuscript, the author also envisaged to list and compare the various methods and their efficiency in a table. However, due to the diversity and complexity of the visual information of harvesting robots, it is difficult to form a unified comparison standard for a large number of research literatures.

For example, it is difficult to find common comparison indicators for different picking objects, different field of view scales, different positioning target points, and different visual characteristics. If the above information is simply listed in the table, may cause one-sided misunderstandings to the readers.

In view of this, this manuscript made a detailed classification according to the research starting points of different literatures, and made an objective summary of the common problems.

 

The above are my main corrections in the paper and the responds to the Reviewer’s comments. Special thanks to you for your comments. We tried our best to improve the manuscript and made some changes in the manuscript. Any revised portion made to the manuscript are marked up using the “Track Changes” function. At the same time, we appreciate for Editors and Reviewers warm work earnestly, and hope that the correction will meet with approval.

 

Once again, thank you very much for your comments and suggestion. If you have any queries, please don’t hesitate to contact me at the address below.

Thank you and best regards.

Yours sincerely,

Qingchun Feng

Author Response File: Author Response.pdf

Reviewer 3 Report

This is a very well written review. It clearly presents all the problems
related to automatic fruit harvesting by robots. The subject is constantly
evolving and new solutions appear, which means that the demand for such
summary work is great.  

 

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled ‘Advance of Target Visual Information Acquisition Technology for Fresh Fruit’s Robotic Harvesting: A Review’ (ID: agronomy-1720546). Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches.

Once again, thank you very much for your comments and suggestion. If you have any queries, please don’t hesitate to contact me at the address below.

Thank you and best regards.

Yours sincerely,

Qingchun Feng

Reviewer 4 Report

This paper provides a review of vision technology for fresh fruit's robotic harvesting. Overall, the paper covers a good coverage of the recent advance of visual technologies. Some of the sub-sections, require more in-depth review, such as 4.1. For example, line 213-214, it was mentioned what Vitzrabin proposed but what's the outcome? The same for line 225-226, what's the outcome? Also, how are weeds related to harvesting (please justify). This applies to section 6.1 as well (line 390 to line 403).

Another to note as well is YOLO. YOLO has many versions, up to v5 but that's not represented in the text (line 248 to line 266). 

Lastly, grammatical check please.

Line 105: (f) needs to be referenced

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled ‘Advance of Target Visual Information Acquisition Technology for Fresh Fruit’s Robotic Harvesting: A Review’ (ID: agronomy-1720546). Those comments (at the bottom of the cover letter) are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval.

The main corrections in the manuscript and the responds to the comments are as following:

Point 1: This paper provides a review of vision technology for fresh fruit's robotic harvesting. Overall, the paper covers a good coverage of the recent advance of visual technologies. Some of the sub-sections, require more in-depth review, such as 4.1. For example, line 213-214, it was mentioned what Vitzrabin proposed but what's the outcome? The same for line 225-226, what's the outcome?

Response 1: We apologies for the negligence of the outcome of the references. The authors have modified the same problem in the full manuscript. For example:

Lines 213-214, the statements of ‘Vitzrabin [69] proposed the adhesive sweet pepper fruit segmentation method according to the depth gradient on fruit.’ was corrected as ‘Vitzrabin [69] proposed the adhesive sweet pepper fruit detection method according to the depth gradient on fruit, which obtained the true-positive rate of 0.909 under the natural illumination.’

Point 2: The same for line 225-226, what's the outcome? Also, how are weeds related to harvesting (please justify).

Response 2:  For the original reference [74], the author's intention is to express the diversity of traditional machine learning algorithms for agricultural object recognition. However, in view of the fact that this document has little relevance to harvesting robots, the authors deleted the document after consideration.

Point 3: This applies to section 6.1 as well (line 390 to line 403).

Response 3:  Considering the Reviewer s suggestion, the authors adjusted the sentence order to summarize the specificity of the passive visual system.

Lines 390-403, the statements of ' The passive detection vision system is simple in configuration and easy to use, but the detection range is limited. It is difficult to cover the fruit distribution area accurately, and the images randomly collected from a fixed perspective will contain much redundant information.' was adjusted to the end of the paragraph.

Point 4: Another to note as well is YOLO. YOLO has many versions, up to v5 but that's not represented in the text (line 248 to line 266).

Response 4:  The authors have modified the same problem in the full manuscript. For each citation to YOLO, the authors have supplemented the corresponding version.

Point 5: Lastly, grammatical check please.

Response 5:  The authors rechecked the manuscript and improved the language wherever possible to increase the readability of the article. Although we didn’t list them in the cover letter, any revised portion made to the manuscript are marked up using the “Track Changes” function.

Point 6: Line 105: (f) needs to be referenced

Response 6: The authors have supplemented citations to the sub-figure 1(f). The contents are as follows:

Figure 1. Typical harvesting robots. (a) Sweet pepper harvesting robot [10]; (b) Strawberry harvesting robot [16]; (c) Tomato harvesting robot [7]; (d) Apple harvesting robot [15]; (e) Apple harvesting robot [20]; (f) Kiwi harvesting robot [18].

 

The above are my main corrections in the paper and the responds to the Reviewer’s comments. Special thanks to you for your comments. We tried our best to improve the manuscript and made some changes in the manuscript. Any revised portion made to the manuscript are marked up using the “Track Changes” function. At the same time, we appreciate for Editors and Reviewers warm work earnestly, and hope that the correction will meet with approval.

 

Once again, thank you very much for your comments and suggestion. If you have any queries, please don’t hesitate to contact me at the address below.

Thank you and best regards.

Yours sincerely,

Qingchun Feng

Author Response File: Author Response.pdf

Reviewer 5 Report

The article represents a nice and comprehensive overview of the never been more interesting and important topic of 'robotic harvesting and computer vision' specialized in highly contrasted fruit and vegetable cropping.

Despite the impression of comprehensiveness, authors are advised to, and they also should, emphasize importance of robot/harvester mechanical solutions, issues and overal structural constitution to cropping/harvesting performance and quality. Mechanical aspect of robotic solutions are not negligeable and in conjunction with appropriate analysis method can give (and be forgiving to some issues) quite good results.

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled ‘Advance of Target Visual Information Acquisition Technology for Fresh Fruit’s Robotic Harvesting: A Review’ (ID: agronomy-1720546). Those comments (at the bottom of the cover letter) are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval.

The main corrections in the manuscript and the responds to the comments are as following:

Point 1: Despite the impression of comprehensiveness, authors are advised to, and they also should, emphasize importance of robot/harvester mechanical solutions, issues and overal structural constitution to cropping/harvesting performance and quality. Mechanical aspect of robotic solutions are not negligeable and in conjunction with appropriate analysis method can give (and be forgiving to some issues) quite good results.  

Response 1: It is really true as Reviewer said that the mechanical aspect of robotic solutions cannot be negligeable. While our topic is focused on the visual information acquisition, we appreciate your suggestions. In the future, we will pay more attention to the progress and trends of the overall system design of robots.

 

The above are my main corrections in the paper and the responds to the Reviewer’s comments. Special thanks to you for your comments. We tried our best to improve the manuscript and made some changes in the manuscript. Any revised portion made to the manuscript are marked up using the “Track Changes” function. At the same time, we appreciate for Editors and Reviewers warm work earnestly, and hope that the correction will meet with approval.

 

Once again, thank you very much for your comments and suggestion. If you have any queries, please don’t hesitate to contact me at the address below.

Thank you and best regards.

Yours sincerely,

Qingchun Feng

Author Response File: Author Response.pdf

Reviewer 6 Report

This paper presents a survey on recent visual information acquisition technology for fruit harvesting robots. The authors also analyzed existing chanllenges and show their perspectives on future research trends. I think this work needs minor revisions. Some suggestions:

1. The title focuses on image acquisition techonology. However, section 4 and 5 discuss about identification and localization. In my opinion, image acquisition is a step prior to image processing, such as detection and localization. The authors may consider modifying the title. 

2. I think visual servoing should be discussed in a specific section.

3. Line 483, typo 'reaches'. 

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled ‘Advance of Target Visual Information Acquisition Technology for Fresh Fruit’s Robotic Harvesting: A Review’ (ID: agronomy-1720546). Those comments (at the bottom of the cover letter) are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval.

The main corrections in the manuscript and the responds to the comments are as following:

Point 1: The title focuses on image acquisition techonology. However, section 4 and 5 discuss about identification and localization. In my opinion, image acquisition is a step prior to image processing, such as detection and localization. The authors may consider modifying the title.

Response 1: Thank you for your suggestion. However, the theme of the title is the Acquisition of Visual Information, which inevitable include a series of contents such as target color, shape and pose, and is not limited to the image itself. Considering Reviewer’s suggestion and the thesis of the article, the authors decided not to modifying the title.

Point 2: I think visual servoing should be discussed in a specific section.

Response 2:  The visual servo applied to the harvesting robot mainly involves two aspects: ① Dynamic servo control when the end-effector is close to the fruit, improving the robot's operation accuracy on the fruit target; ② Adjusting the camera attitude servo control according to the spatial distribution of the fruit target , to achieve the purpose of active search.

  Since the theme of this paper is the Acquisition of Visual Information, the visual servo involved in ① focuses more on the control of the robotic arm's executive components, rather than the process of visual information acquisition itself. Therefore, this paper mainly reviews the servo control methods related to ② in section 6.2. Active Detection with Multiple View Field.

Point 3: Line 483, typo 'reaches'.

Response 3:  Line 483, the statements of ‘In the current reaches' was corrected as ‘In the current research’.

 

The above are my main corrections in the paper and the responds to the Reviewer’s comments. Special thanks to you for your comments. We tried our best to improve the manuscript and made some changes in the manuscript. Any revised portion made to the manuscript are marked up using the “Track Changes” function. At the same time, we appreciate for Editors and Reviewers warm work earnestly, and hope that the correction will meet with approval.

 

Once again, thank you very much for your comments and suggestion. If you have any queries, please don’t hesitate to contact me at the address below.

Thank you and best regards.

Yours sincerely,

Qingchun Feng

Author Response File: Author Response.pdf