Displacement Measurement Based on UAV Images Using SURF-Enhanced Camera Calibration Algorithm
Round 1
Reviewer 1 Report
Detailed Comments:
1. Displacement measurement is not a new problem, and there are many research methods. What are the problems that this institute aims to solve? What is the innovation point? The article does not give clear instructions.
2. The methods presented in this study are some of the existing mature methods, what are the main contributions of the authors? It needs to be carefully summarized.
3. If it is a new method, it needs to be compared with other existing advanced methods and its advantages need to be discussed in detail.
4. The final analysis and discussion in the article is too simplistic and does not meet the requirements of writing a research paper.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
In my opinion, this work is very interesting because presents a first step towards the use of UAV video for infrastructure displacement measurement as a low-cost technique. The problem with using UAV imagery is that the camera movements are coupled in the movement of the target and decrease the accuracy of the displacement measurement. The proposed methodology uses the digital image correlation method, but before, the image taken by the mobile camera is corrected by SURF (speeded-up robust features) feature point tracking and the MSAC (m-estimator sample consensus) algorithm. The work shows results obtained by numerical simulation, with an experimental verification. Finally, influence of different wobble modes of the UAV on the measurement results are discussed.
The paper presents a good bibliographic review, with high impact citations.
However, the manuscript is not yet ready for publication, but it is very promising if the authors are willing to make the following small minor revisions
- Line 384 introduces the acronyms MMAE and MRMS, and does not indicate their meaning. Previously only the meaning of MAE and RMS has been indicated.
- Figure 9 has to be redrawn: truth value displacement time-history is not distinguishable
- It is recommended that the meaning of ROV and FROI be indicated in Conclusions.
- Figure 13 can be shown at a larger size, so that the different displacement time-history curves can be better distinguished.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Minor changes in English such as in line: 112, 119 (series of images); 62 (LDV or LVD); 269 (Virtual) apart from that a thorough check of article is suggested etc.
Overall composition of Article is well formulated and research component is fine. A justification of the selected algorithms for the study might improve the novelty. Concluding remarks suggests high calculation accuracy and stability to image distortion, this must be tested in multiple experiment scenarios as compared to provided synthetic environment.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
I have no other comments.
