EasyIDP: A Python Package for Intermediate Data Processing in UAV-Based Plant Phenotyping

Round 1

Reviewer 1 Report

I am positive to the manuscript includes a lot of valuable information for high-throughput phenotyping. And also, authors evaluated well in different growth type of crops in research field. Although I would recommend acceptance for publication, it should be better authors arrange to consist of this paper.

In my opinion, the manuscript is too long, it would be probably better describe along with the title scope.

• As for ‘Abstract’, it seems not to be described well this manuscript.
• As for ‘introduction’ & ‘discussion’, I felt redundant as for description like general instruction book. It should be better focus on the reason why you would consider to construct your package as for their efficiency and accuracy.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have developed a Python package named EasyIDP which automate some of the intermediate data processing tasks required in UAV-based plant phenotyping tasks. This manuscript describes the principles behind the functions introduced in the package and discusses some prospects of the package. However, the manuscript needs much more grammatical attention. I have marked the grammatical mistakes whichever that was obvious, but other than that the authors should check thoroughly the sentences so the readers can understand. I also suggest updating the quality of all figures. Upon the proper updates, the manuscript can be considered for review again.

Major Comments:

1. Line 33 – Line 38: The introduction is considered as a place for providing high-level information on the problem and maybe the existing solutions. I suggest the authors to remove the lines where the breeder’s equation is provided and discussed. I would simply point out the importance of high-throughput non-destructive phenotyping using some citations. This may discourage a reader from getting the brief idea about the problem.
2. I have found many sentences need proper grammatical corrections. For instance, sentences were made lengthy using complex components, which can be easily described in small sentences. This will really help the readers to follow along the procedures. I have tried to mark such sentences in the Minor Comments section, but please check the whole manuscript one more time.
3. The Figures in thin manuscript definitely require updates. First of all the resolution of the figures are very poor. I cannot understand anything from the figure clearly other than some blurry objects. Also, the orientation of the Figure 3 can be changed into landscape mode to utilize all the blank spaces in the page. Since Figure 3 and Figure 4 are very important in visualizing the results, it is not possible to comment on the explanations without clear figures.

Minor Comments:

1. Line 3: Remove the full stop from the end of the title.
2. Line 32: Too many nouns in one sentence. Consider replacing the sentence like this: “… the increase rate of current yield gains needs to be doubled by 2050”.
3. Line 40: Add “which is” before “not only time-consuming”
4. Line 42 – 44: Consider rewriting the sentence as the previous sentence also stated “not only … but also”.
5. Line 46 – 49: The sentence should be restructured, may be dividing into two separate sentences? Something missing before the “…normally difficult 3D related structural traits…”.
6. Line 51: Please consider defining the acronym UAV here since it is mentioned here first.
7. Line 61: At the end, may be reference the Figure 1?
8. Line 89: “automatic” should be “automate”.
9. Line 89: Figure A1 or 1A?
10. Line 89 – 91: The whole sentence looks problematic. Please change the wordings.
11. Line 103: Just use machine learning, because deep learning can be considered within machine learning.
12. Line 103 – 104: The sentence needs to be restructured. Hard to read.
13. Line 104 – 105: Shouldn’t it be “…neither a detailed computation algorithm”?
14. Line 107: Missing full stop (.).
15. Line 114: Should be “various” crops.
16. Line 117: “developing” should be “develop” and “testing” should be “test”. Note that its always present tense after “to”.
17. Line 125: Pleas capitalize the first letters of each crop in Table 1.
18. Line 129: Here authors mentioned “flight height of 30/20 m”, but in the table there are heights of 30m and 50m only. Please clarify.
19. Line 127 – 131: If you have, please insert the pictures of the two systems that you collected in the field.
20. Line 132: 2.23D? Shouldn’t it be 2.2?
21. Line 133: What device? I guess it was a computer with all those configurations. Please move this information at the end of the section 2.2 Reconstruction by SfM-MVS. Also, replace “including”, this does not seem right in this sentence.
22. Line 137: “to processing” > “to process”
23. Line 140: Be consistent with the term. Use either UAV or drone. I suggest using UAV.
24. Line 143 – 145: The whole sentence needs to be rewritten. Too many parts and too many missing prepositions.
25. Line 145 – 151: The rest of the paragraph also has major grammatical problems. For example, you said, “After finishing processing align photos, check the derived camera height…”. This should be like this: “After processing the image alignment operation, the derived camera height was checked. If the derived height was not close to the actual flight height, the image heights were manually modified…..”. Please revise the whole paragraph.
26. Line 159: “For produced” or “to produce”?
27. Line 167: At least mention the Python package names and use proper citation. Since this manuscript is about a high-level software, you should detail these parts.
28. Line 169: Capitalize first letter of cropping.
29. Line 197 – 200: Very hard to read. Missing a lot of components and too large sentence.
30. Line 204: At least mention the Python package names and use proper citation.
31. Line 211: include or including?
32. Line 217 – 219: The second part of the sentence is confusing. Revise.
33. Line 237: “it makes ….” > “which makes”
34. Equation 1: Why the T matrix has only two components in the 1^st row (only R and t), whereas clearly the matrix shape is 2,4?
35. Line 267 – 271: Again, too much information was tried to put in one sentence and too many missing components were there.
36. Figure 3: Remove the numbers on front the crop types. Also capitalize the first letter of the Crop names.
37. Figure 4: Change the crop names according to Minor Comment#35. Also, the readers are not interested in the filenames. Remove the filenames so the figure does not look cluttered. Also I don’t see any black dash line which was shown in legend as Pix4D. Please explain.
38. Line 345: “manually drawn”.
39. Line 349: “…first kind of comparison were shown…” Please check your tenses. The methods should be in past tense which you did, but the figure which you are showing now (in the manuscript) should be in present tense.
40. Line 355: “By manually picked…” > “By manually picking…”
41. Line 363: “reverser” or “reverse”?
42. Line 378 – 379: “… once the pixel distance to the image center smaller than 800…” this sentence is missing a verb, i.e., “is”.
43. Line 380: It is “to minimize”.
44. Line 381: “…raw images where locates in the center were recommended” grammatically this sentence is wrong. Even so, please explain clearly what are you recommending exactly?
45. Line 386: In Figure 6 caption, please use 5^th percentile and 95^th Please follow this for the entire manuscript wherever you disclose percentile value.
46. Line 389 – 391: Please break down the sentence. It is hard to read.
47. Line 397: “for the central manually marked for all 112 lotus plots”, central manually marked what? If you mean “central manually marked” 112 lotus plots, then revise the sentence. In its present form, it seems there is something missing.
48. Line 413: “time-serious”?
49. Line 413 – 415: Break down the sentence.
50. Line 417 – 419: The sentence is grammatically wrong. You cannot just add two different things by just using a comma (,). Please revise it.
51. Line 429 – 433: The entire sentence is 5 lines long, very hard to read. Please break it down so readers can track the information easily.
52. Line 441 – 442: “… it is quite difficult to find dataset (that) suits specific phenotyping tasks”, missing “that”.
53. Figure 8: Please reduce the number of images so that each subset image is clearer. In its current form, even you increase the resolution, it is hard to understand the differences. Also, remove the filenames with .JPG on it.
54. Figure 9: Same comment as Figure 8.
55. Line 484: isn’t > is not.
56. Line 491: already labeled what? Plot?
57. Line 493 – 495: The citations are broken. Please correct. Also define GAN as Generative Adversarial Networks. Not everyone know this acronym.
58. Line 499: not “were”, it should be “are” since this is the current state of the software.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

13/06/2021

Dear authors,

 

In the manuscript EasyIDP: A python package for intermediate data processing in UAV-based plant phenotyping you propose an open-source software package called EasyIDP (easy intermediate data processor, MIT License) which you have developed to decrease the workload in intermediate data processing. The objective of this study is to develop a lite and easy-to-use software package to address previously identified difficulties and decrease the workload in intermediate data processing for agronomists, including 1) cropping both large PCD and DOM to small parts by given ROI without reading the whole file into memory at once; 2) reverse calculating given ROI on the corresponding place on raw UAV images for both Pix4D and Agisoft Metashape; 3) testing the accuracy and performance of the developed functions using six different plots with variant crops.

General comments

The theme is wary actual and interesting. My personal assessment on the work is positive, owing to the relevant, timely and clarity of the study, as well as the focus of analysis.  The study is interesting and results may have great usability in the domains of photogrammetry and remote sensing, and also agriculture.

However, the figure from 3 to 9 is very poor quality and need to be replaced. Furthermore, Conclusion chapter is too general. Conclusion should be a detailed final interpretation of all the results obtained in the research, and highlight the novelty you introduce. Revised the Conclusion and substantiate the general statements with facts from the results.

Please, read all comments and act on them.

 

Specific comments (are in the manuscript)

• Line 108-19 – This sentence should be reworked, because this is not study but manuscript about your study and result. For example: "The objective of this manuscript is to proposed developed a lite ....”, or something like that.
• Line 317 – Figure 3 is very poor quality, image itself and text within picture.
• Line 322 – Is this misspelled? Where is Figure A1?
• Line 342 - Figure 4 is very poor quality, image itself and text within picture.
• Line 365 - Figure 5 is very poor quality, image itself and text within picture.
• Line 385 - Figure 6 is very poor quality, image itself and text within picture.
• Line 394 - Is this misspelled? Where is Figure A2?
• Line 397 - Figure 7 is very poor quality, image itself and text within picture.
• Line 474 - Figure 8 is very poor quality, image itself and text within picture.
• Line 480 - Figure 9 is very poor quality, image itself and text within picture.
• Line 508-518 - Conclusion chapter is too general. Conclusion should be a detailed final interpretation of all the results obtained in the research, and highlight the novelty you introduce.

 

Best regards

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf