A Probabilistic Approach for Stereo 3D Point Cloud Reconstruction from Airborne Single-Channel Multi-Aspect SAR Image Sequences

Round 1

Reviewer 1 Report

The manuscript “A Probabilistic Approach for Stereo 3-D Point Cloud Reconstruction from Airborne Single-Channel Multi-Aspect SAR Image Sequences”, authors Hanqing Zhang, Yun Lin, Fei Teng, Wen Hong, investigates the problem related to a dense 3-D point cloud generation by means of stereo radargrammetry using multi-aspect SAR image sequences: they used a mixture distribution model for the stereo matching results, where scattering anisotropy was modeled as an independent error source, proposing a Bayesian filtering method for the generation of dense 3-D point clouds; authors tested the proposed method on a Ku-band drone SAR data for Taiyuan Yaocheng Airport of China.

 

In my opinion, authors present an interesting work for the readers of Remote Sensing journal.

 

The biggest problem that I found in this manuscript is topicality. I noticed that there are 37 References, but only 4 were published after 2018: moreover, 2 of them are self-citations: in addition, they compare the proposed method with a method published 10 years ago… I remind the authors that the topicality of a publication is a crucial aspect… for these reasons I encourage the authors to make an effort to update the References.

 

Figures need to be improved by enlarging the labels for better readability.

 

All acronyms must be defined when written for the first time (please check some acronyms… e.g. CSAR – line 108,…).

 

Authors do not provide motivations about the chosen values of some parameters in their method (see page 10, lines 297-298, line 304, α = 3, page 12, line 351): please, provide reasons for these choices…

 

Please, specify the “Ground Coordinate System (GCS)” in the experiment: local? Global? UTM reference system?

 

 

In the Discussion section the results are not discussed: the authors summarize the manuscript, with some conclusions; usually, in this section the results obtained should be compared with those of other authors: I suggest to entitled section 5. “Conclusions”.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors of the manuscript propose propose a mixture distribution model for the stereo matching of consecutive multi-aspect SAR images, where scattering anistropy one of the main problems to be resolved and it is modeled as an independent error source, followed by a Bayesian filtering method the generation of dense 3-D point clouds.

The manuscript and work behind is very well presented and justified by comparing the proposed approach with other from the litarature. The results depict an improvement in the final output.

Among small typos and proofreading that is required (e.g., writting a 3D is not consistent), my main concern is with the Discussion section, which i would call it in this form a conclusion. It would be a better ending for the manuscript if the discussion is better developed (by discussing the results by themselves, compared with other litarature, discussing the benefits of applying the approach in real case scenarios/applications, etc.). Not just briefly reiterating the work done.

 

Regards.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Point 1: The authors have to improve their literature discussion. Not much Recent literature has been given in this manuscript. Please add more recent works of literature (2018 onwards). Add a discussion to critically state research gap and how the proposed findings have addressed the research completed.

 

Point 2: Please re-standardize the abbreviations in the manuscript, e.g. "CSAR", "MASAR", "CAD" are unclear in their meaning.

 

Point 3: Line 380, “4.2.1. Comparative Evaluations” should be deleted.

 

Point 4: Please, rewrite Discussion providing clearer information on the results obtained. Also, describe in Discussion where the results obtained can be used and further research perspective.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

From a technical point of view this manuscript is presented in an attractive manner. The authors seek to achieve high-precision 3D reconstruction of urban scenes with only stereo radar measurements. Experiments on real dataset are quite impressive and have the potential to be published in remote sensing. But I have some observations.

 1) The authors need to improve their discussion section. In my opinion, the discussion section in this manuscript acts more like a conclusion. The author may directly change this subtitle to “Conclusion”, or give a more standard “Discussion”.

 2) I also recommend authors to update their references. The reference should be made to more recent literatures. To this end, authors may also refer to some broader studies on their topics, and reduce the amount of ‘outdated’ references.

 Minor :

3) Abbreviations should be defined the first time they appear in the manuscript. Such problems includes：’MASAR’ in ‘Line 195’,  ‘CSAR’ in ‘Line 108’.

 4) The term ‘ground coordinate system’ may not be suitable used in ‘Line 112’, as this term usually refers to the state plane or UTM grid system rather than some general Cartesian Coordinate systems. Authors should be careful with the use of such terms.

 5) I must remind authors that if they employ some tuning parameters in their algorithm, they have to explicitly point it out, instead of just offering the value of such parameters. An example is,’‘ in ‘Line 296’.

 

6) Line 380：subtitle ’4.2.1’ can be removed from this manuscript.

 

7) The text in Figure 8 should be enlarged to make it easier to read.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

In this second version of the manuscript “A Probabilistic Approach for Stereo 3-D Point Cloud Reconstruction from Airborne Single-Channel Multi-Aspect SAR Image Sequences”, the Authors have improved the work addressing most of my requests / comments: I suggest publication in the present form.

Reviewer 4 Report

It's a pleasure to accept the paper.