MISF: A Method for Measurement of Standing Tree Size via Multi-Vision Image Segmentation and Coordinate Fusion
Marcos Benedito Schimalski
Round 1
Reviewer 1 Report
The manuscript is interesting for an alternative approach to segmentation of a single tree by a combined method, which is based on multi-vision image segmentation and coordinate fusion to realize the accurate recognition and segmentation of the main body of standing trees.
There are some recommendations for improvement:
1. It is appropriate to add two sentences at the end of the abstract that answer the questions: "What do the results mean in practice? And what remains unresolved?"
2. Please explain the poor quality of figures 7, 8 and 9. The quality should be improved
3. For a more complete demonstration of all the advantages of the experiment, it is desirable to supplement the Introduction section with information on the practical application of the results. I recommend in the introduction to clearly define the number of tasks, and
4. To increase reader interest, it is advisable to disclose future research at the end of the Discussion section.
5. It is desirable to include in the keywords the Latin name of the tree (or trees), the individual segmentation of which was studied in the research using the combined method. Also in the text of the manuscript, when mentioning the name of a plant species, it is necessary to give the Latin name of the taxon in parentheses.
6. What is MISF in the name? This abbreviation is not disclosed anywhere in the text.
7. In conclusion, it is necessary to briefly and clearly summarize the achievement of the three tasks indicated in the introduction.
8. In conclusion, it is necessary to remove the inappropriate statement and provide specific numerical values of the data obtained with an indication of the p level.
9. It is desirable to specify DOI in the references.
10. The Data Availability Statement section is missing. Where can I see the results (datasets) of these studies? You must provide a link to them.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Authors:
Dear Authors:
Your article: “MISF: A Technique for Measurement of Standing Tree Size using Multi-vision Imaging Segmentation and Coordinate Fusion" presents a relevant topic that still has gaps to be solved. Your work contributes precisely by developing an automatic methodology for the extraction of dendrometric variables from individual trees. In order to contribute to the improvement of the article and the publication of its final version, I request you to carry out a detailed review of the final text, observing concordance, tense, and meaning. When presenting an acronym for the first time in the text (SURF – line 61, for example) please present its meaning. I have submitted 28 comments of lesser relevance and which can be viewed in the digital file. I request your special attention to the following observations:
1) The abstract gives the idea that the experiment was carried out in a stand. Please review and highlight that the tests were implemented on isolated trees in an urban environment.
2) Lines 35 to 38: please better detail the methods and sensors used.
3) Lines 44 to 52: you present bibliographical references that used photogrammetric methods to obtain dendrometric variables. Only 3 references are presented and one of them [15] with monocular vision. It would be interesting to cite a few more studies.
4) Lines 61 to 70: highlight the objectives to be achieved in this part of the article.
5) Lines 73 to 75: “ In the theory of close-range photogrammetry, using a camera to capture images to obtain target information is called monocular measurement, which is low cost, but it can only obtain two-dimensional information from the target, and cannot determine the distance of the measured target.” There is a little confusion here. In the concept of monocular vision, only one image is used for measurements. This same camera can be used to obtain images with overlay and then extract stereoscopic information.
6) Please detail the camera used, as well as the hardware configurations. Appropriately cite the applications used.
7) Cite the bibliographic reference for the use of the Levenberg-Marquardt method;
8) Lines 226 to 223: present a figure to illustrate the image acquisition method.
9) Line 249: “The color image needs to be grayscaled”. Please cite the mathematical model used in this conversion.
10) Line 262: “Finally, the 112-dimensional descriptor composed of the 64-dimensional feature descriptor obtained by the SURF algorithm and the 48-dimensional color descriptor”. Please, cite the bibliographic reference for the query of the mentioned descriptors.
11) Detail the experiment (chapter 2.5.1): location, tree species measured, distance from the camera to the trees (distance/height ratio), approximate angles of the images taken, and total number of images.
12) Line 274: “Measuring the size attributes of a standing tree using close-range photography requires the integration of the geographic coordinate system and the camera coordinate system, rather than a certain geographic coordinate system”. In fact, a terrestrial reference system is needed, which can be global or local. Please rephrase the idea at this point.
13) Equation (33) presents the calculation of the three-dimensional distance between two points in space. Does this value correspond to the height of the tree? Please review.
14) Illustrate the variables measured in chapter 2.4.3. Use the products generated by its processing as a reference for this illustration.
15) Will the developed algorithms be made available on GitHub?
16) Line 408:” Another method using UAV images”. The same parameters were used: flight height, and longitudinal/lateral coverage cited in the bibliographic reference.
17) Line 411: “is compared with that of another method using a single image taken by a mobile phone”. Please detail the procedure used.
18) Line 429: “The focal length of the three cameras is 27mm.”. In this part of the text, it became clear that three cameras were used simultaneously. For each tree were only three images obtained? Please provide more detail on this step and the arrangement of cameras for recording images.
19) Line 448: Was the evaluation of the segmentation results only visual?
20) The results presented in Tables 1, 2, and 3 would be easier to read by highlighting (bold) the best results obtained.
21) For the results presented in tables 1, 2, and 3: are the biggest errors related to the biggest measurements? Are the smallest errors related to the smallest dimensions?
22) Figures 8a and 8b: we have exactly the same images.
23) Normally, the size of the crown of a tree is not uniform. You used a preferred direction to measure the crown width.
24) Conclusions: add the conclusions: a) Were the objectives proposed for the study duly achieved? b) recommendations for future studies.
I conclude by congratulating them for the work done and for the presented version of the article.
Respectfully,
Comments for author File: 
Comments.pdf
The final draft must be sent to a native English speaker or a reputable proofreading agency for editing.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Dear Authors,
The new version of your manuscript entitled "MISF： A Method for Measurement of Standing Tree Size via Multi-vision Image Segmentation and Coordinate Fusion" presented several changes that made it more fluid to read. With this, the understanding of the experiment and interpretation of the results was excellent.
I carefully revised your entire manuscript and facilitated this work with the help of the cover letter.
Congratulations on the work and manuscript presented.
Respectfully,
Dear Authors,
About the writing of the text and I could see that a more detailed review is needed, especially with regard to the use of commas in your sentences.
Overall the text is well-written and only needs these minor tweaks.
Sincerely,
