Integrating Multi-Point Geostatistics, Machine Learning, and Image Correlation for Characterizing Positional Errors in Remote-Sensing Images of High Spatial Resolution

Round 1

Reviewer 1 Report

Dear author,

Thank you for your submission. The paper is indeed promising but could benefit from some revisions to enhance its understandability and overall impact. Here are detailed suggestions:

 

1. Please ensure all abbreviations are fully explained at their first use in both the main text and within the figures or tables. This will improve readability greatly. This includes terms such as MPS, TIN, and SGSIM.

 

2. In section 2, while the written description is informative, it would be more beneficial and meaningful if you could incorporate key formula expressions related to the methodology.

 

3. The concept of "positional errors" needs a clear definition. Please provide this for the reader. Similarly, the covariable in Section 3.2 also needs a definition.

 

4. Figure (b) and (c) require better clarity. Unless the improvements significantly enhance understanding of the text, consider removing Figure (a) as suggested.

 

5. In line 149, there is a need to clarify what is meant by "the reference sample data are considered as true values." Make sure it is apparent that these are the true values of the images, not the positional errors.

 

6. In Table 1, it is unclear how the same points between the reference and the test image are matched. Please provide an explanation of the technique used.

 

7. You have mentioned different detrending methods resulting in various residuals, yet only selected the GAM method. Please provide an explanation why you chose this particular method.

 

8. Finally, in section 3.3, a comparison among the three Stochastic simulation methods (TIN, SGSIM, and DS) would be useful. This will provide readers with a clear understanding of the distinctions and applications of these methods.

 

Once these recommendations are addressed, please also consider reviewing the overall English language and flow of the paper. Proper grammar, syntax, and consistency in terminology usage can greatly enhance readability.

 

Looking forward to the revised version of your paper. 

Once these recommendations are addressed, please also consider reviewing the overall English language and flow of the paper. Proper grammar, syntax, and consistency in terminology usage can greatly enhance readability.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript presents authors’ work on positional uncertainty in remotely sensed images of high spatial resolution. It promotes combined use of DS, GAM, and computer image correlation for geostatistcal simulation of positional errors. The whole manuscript seems good, but there are still some issues to be addressed. 

1. Figure 1 is too large. Too much unnecessary details are listed. Some parts of the part is obscure. For example, the first two interwined arrows. It is hard to judge how it proceeds. Please simplify this figure.

2. The spatial distribution of training data is too regular. How were the real values of the training data detemined? And this regular distribution may reduce the usability of your model in real life applications.

Author Response

Please see the attachment.

Author Response File: Author Response.docx