A Method for Detecting Feature-Sparse Regions and Matching Enhancement
Round 1
Reviewer 1 Report
This was an interesting paper that presented a sound concept for improved multi-modal image matching. I would recommend a few minor improvements/corrections to help better present the material: Line 56: Non-linear radiation distortions are mentioned but not explained at all beyond a reference. Given the focus of the paper, it would be helpful to specifically discuss the cause of the distortions and how it impacts matching algorithms. Line 264: The last statement should reference Fig 6d I think. Figure 7: Did you consider increasing the sampling of your trials to include S=128 and 512? It seems like there is a big gap in results between 256 and 64 or 1024 so I wonder what the results would be with increased sampling in that region. Figure 9: Is the intent here to show that with lower thresholds more features would have been detected? That isn't apparent. Table 1: Would it be possible to show the number of correct matching points as a percentage of the total matching points (including both correct and incorrect points)? Line 391 references the accuracy of the ContextDesc algorithm but this is not shown and would be useful data to evaluate the algorithms.Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript proposes an algorithm (SD-ME) to solve the matching problem between UAV images and satellite images. Compared with other algorithms, the authors get better results. Overall, it may bring some value in image matching.
1. The author should revise the English representation of the abstract in the manuscript.
2. It is suggested that the author write results and discussion separately and standardize the title naming, which is helpful to improve the structure and legibility of the article.
3. Please set an appropriate font size for Figure 7 and Figure 11 to increase clarity.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
With interest, I read the manuscript. It is appreciated that the manuscript is easy to follow and not too long. The message is clear and of interest to the community. The authors proposed a paper titled "A Method for Detecting Feature-Sparse Regions and Matching Enhancement". The proposed paper seemed to be promising in terms of computational simplicity and classification accuracy. I would accept it in this present form.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
