Fusion of Hyperspectral and Multispectral Images with Radiance Extreme Area Compensation
, Geng Zhang 1, Shuang Wang 1, Siyuan Li 1 and Yupeng Liu 5Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis paper presents a method for dealing with over/under exposed areas of the signal during multi-spectral pan sharpening, or increasing the resolution of hyper spectral bands using less spectrally rich information. A problem that is several decades old.
The method is an incremental improvement on existing algorithms that nevertheless deals with an important issue: regions where signal is essentially lost. The proposed solution to look for similar areas around the signal in order to provide constraints on the solution is a sound one even if it rings of in-painting and may never truly recover the signal. The key is that the signal lost to overexposure say may not be similar to its surroundings, but there is no other reasonable approach. So the method is not groundbreaking but is reasonable.
Local smoothness constraints need to be explained better. There is no reference to figure 5 or what it shows.
What can be improved the most is evaluation: why didn't you reduce the resolution of the true satellite HS/MS data so you have a true reference HH, even if it is at less resolution than what is available. This should be done.
Also it is not clear how many scenes you cut out of the first two datasets. Since you use smaller sections. Is it only one or several. It is not thus clear if the quoted scores in the tables are for a judiciously chosen section where the method works best. This needs to be stated and results on as much data as possible provided.
Finally, final sentence on page 16 makes no sense. Check grammar.
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Authors, I've read your paper that I've found very interesting and well organized. I list some basic sugestions:
- In the introduction a relevant number of citation is provide an they are well presentated in your paper framework. My only sugestion is to move figure 1 in the result section because you are presenting, in the introduction, an initial result and discussion of your research (" we have found that the fusion quality is poor in some extreme radiance regions by comparing fused hyperspectral data with reference data and calculating the spectral angle error, as shown in Figure 1."). Is my opinion that you could leave the sentence here and move the figure in other section...also because this kind of consideration/conclusion need more attention and data to support your thesis (introduction is not the right place).
-line 172: "section % summarizes the paper" I suggest to substitue "summarize" with "conclusion"...section 5 is conclusion (is not a résumé).
- Maybe from lines 140 to 157 can be moved at the beginning of section 2- methods
-Add a table with the dataset used in the paper
- Can you better explain the use of SAM for the error map production?
-All long the paper be more precise to explain acronyms. When a formula is presented mathematical terms must be explained. please check it.
-The validation process could be enhanced and better described
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe problems in the paper are summarized as follows:
1. It is suggested to further concretize the contribution, rather than just providing a title and theoretical support, highlighting the scientific value of this contribution.
2.Pay attention to the consistency of symbols in the paper. The writing form of R in line 187 is inconsistent with that in formula 2, and consistency needs to be maintained.
3.The process in Figure 2 is relatively simple. It is recommended that the author refine the improvements or innovations made, so that readers can understand the differences and main contributions of this article compared to previous work.
4.What are the physical meanings of the horizontal and vertical coordinates in Figure 5? Suggest marking it in the diagram.
5.There are a lot of writing errors in the text, for example, Eh and Em in Table 1 are inconsistent with the main text where h and m are subscripts in formula（15）and （16）.
6.The discussion part should include a portion of the experiments, such as verifying important parameters in the experiment, rather than a purely textual description.
7.The conclusion section should be the improvement achieved in this article, including some quantitative results, rather than a description of the algorithm implementation process. It is recommended that the author improve and modify it.
The method in this article has made some improvements, but there are many typos, and the writing of various parts deviates from the standards of scientific papers. It is recommended to further improve.
Comments on the Quality of English LanguageMinor editing of English language required.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsAccept in present form
Comments on the Quality of English LanguageMinor editing of English language required
