SAR Image Fusion Classification Based on the Decision-Level Combination of Multi-Band Information
Round 1
Reviewer 1 Report
no more comments.
Reviewer 2 Report
I think there are no other problems with the revised version of this article and suggest it be published.
Reviewer 3 Report
Dear authors,
You have made a great work. Congratulations.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
See attached file.
Comments for author File: 
Comments.pdf
Reviewer 2 Report
-
The authors proposed a SAR image fusion classification method based on multi-band information decision level combination is proposed. Combining the conflict coefficient and the idea of TF-IDF, the weights of different sensors are obtained, and the final feature types are obtained by decision-level fusion. The feasibility of the method is verified by two groups of multi-band SAR images for classification experiments. In a whole, this paper can be accepted.
There are some minor problems that need to be revised.
- Why A is chosen as the comparison method, what are the characteristics of this method when classifying multi-band SAR images?
- Are the label samples and test samples consistent when classifying images of different bands?
- Some pictures in the article are not clear.
Reviewer 3 Report
Dear authors,
After a carefully reading of your work I have seen a number of flaws that can be drastically improved.
In general, the work needs for a major grammar editing and style. The number for the references starts very well and from the line 39 to the end they are bad written. I do not know what the authors have made with the numbering of the different section, but it denotes careless. Finally, I am not convinced at all that the proposed method be better than older ones.
More in detail:
Abstract: Some more specific comments on the results must be included.
Introduction:
Lines 30-35: Too long sentence and difficult to understand. Please, write it again.
Line 39: [13] instead of 13
Line 42: idem with 14.
Line 43: A novel reader does not know what a super pixel is. It is effortless writting the sentence in a more readable way.
Line 44: It is hard to understand. The processed data or data processing?
Line 46: See comments for lines 29 and 42.
Line 49: idem line 46
Line 51: idem line 49 and so on ..... The authors must correct them!
Line 78: Section 2.2 after Section 1? Please, this is careless.
Line 78: Why convolution is capitalized and not neural networks?
Lines 84/85: Convolutional layers or Convolution layers?
Lines 91-94: Too long sentence
Lines 140...: The authors take as a very basic of the work the belief entropy, function of the belief degree.
I know a number of ways to compute the entropy but this needs to be clarify
These concepts must be clarified .
Images: In a general way, the images must have coordinates. It is very convenient to say the covered area. How many gray levels (8-bits, 12 bits) have the images? Are the same for all?
Results when computing the 1.7% of improvement in not a significant if the computation effort if greater than the previous technique.
So, what is the real contribution of the work. This must be fully explained and clarified.
I hope these comments be useful to improve your work.
Good luck
