Unmanned Aerial Vehicle–Light Detection and Ranging-Based Individual Tree Segmentation in Eucalyptus spp. Forests: Performance and Sensitivity

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1.     In the introduction, particularly the final paragraph, it is preferable to highlight the novelty of your research.

2.     Kindly provide the reference for Equation 1.

3.     Kindly elaborate on the algorithms described in subsections 2.3.1, 2.3.3, and 2.3.4.

4.     With varying Gaussian smoothing factors, the findings for different plots in Figure 4 are not always the same. For instance, the results on Plot 10 in Figure 4c are precisely the same, but not in the other cases. Would you kindly elaborate on the discussion? Please also explain Figures 8 and 10 using the same equation background.

 

5.     It would be more effective if you could restate a novel finding from your research in the conclusion section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Generally, the paper is well-written and well-structured. However, the method is not particularly novel; it's quite common. Nonetheless, testing in a dense plantation study site can be a positive point. The authors utilized various segmentation methods and considered identifying the best parameters based on forest structure, which are other positive aspects. Nevertheless, the paper needs a bit more work before publishing. Therefore, I have some suggestions:

1. Lines 63-76: Include previous research and provide examples for all approaches. Discuss their strengths and weaknesses.

2. Lines 77-92: This part need more examples, references and explanation.

3. Lines 94-96 to include various studies that have tested several segmentation methods. Discuss their results and then emphasize the novel aspects and assumptions of your research.

4.Figure 1: The plots are difficult to discern. Please increase the zoom and enhance the quality of this section.

5. Lines 143–145 explain how the distance thresholding method was used to remove outliers. How do you determine this threshold? How about the filtering method?

6. Report the unit in Lines 175-176. How did you determine these ranges?

7. How did you utilize the relationship between crown size and height to detect the position of the canopy? Could you provide further elaboration? Additionally, how did you take into account the forest type and crown shape?

8. In section 2.4, what do you mean by 'r' and 'p'? Are you referring to sensitivity and specificity? This section needs revision. Additionally, 'F' does not represent the overall accuracy; it denotes the F-score, which is a weighted average of sensitivity and specificity.

9. The quality of all figures, especially Figures 2 and 3, is not acceptable. It's challenging to discern the details.

10. In Lines 272-273, what exactly do you mean by detection rate, accuracy, and overall accuracy? In Lines 235 and 237, you mentioned that 'r' represents the recall rate and denotes detection rate (omission??), while 'p' stands for precision rate and signifies detection accuracy (commission??). Additionally, 'F' is the scoring rate representing overall accuracy as it comprises a weighted average of detection rate and detection accuracy. However, this explanation seems incorrect and confusing. The accuracy assessment section, along with all relevant parts in the paper, needs to be revised for clarity and accuracy. Also see lines 325-327, 319-320, ...!!!!!

11. what do you mean in this section in the discussion. clarify how they differ between low density and high density plots and why?: The overall value of the detection rate (r) of the WA, LMA, EDCA and LSA methods was 0.729, 0.721, 0.740, 0.672 separately, which indicated that there was no significant difference in the detection results of the four methods. However, the difference between the low-density plots and high-density plots was obvious.

 

Comments on the Quality of English Language

Further revisions are necessary to enhance the clarity and comprehensibility of the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

In this paper, an individual tree detection based on LiDAR data is presented. Topic is interesting and the manuscript is well-written. I just mention some points to improve its quality:

-          There are some grammatical errors such as Where after equations.

-          Figure2: it is necessary to present some more descriptions regarding the used method.

-          Please improve the quality of Figure 3.

-          Why progressive morphological filter is used?

-          Please present point cloud, DTM, DSM of the study area in one figure.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed all my comments for this paper and answered the technical questions. It is ready for publication

Reviewer 3 Report

Comments and Suggestions for Authors

Accept as it is.