An Unsupervised Port Detection Method in Polarimetric SAR Images Based on Three-Component Decomposition and Multi-Scale Thresholding Segmentation

Round 1

Reviewer 1 Report

Although the authors have addressed a lot of the points and improved the paper. Still, they did not address the major issue of novelty. The method is based on the one of oldest segmentation techniques, namely thresholding of histograms.

Also, the paper still needs professional editing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The ms was significantly improved with respect to the previous version according to reviewers' comments. Only comment 3 can be better addressed by including the most significant quantitative findings, i. e., the average time processing and IoU values achieved in the experiments, in the abstract and conclusions. Hence, I recommend the publication as it is. Please just completely revise the ms to avoid typos, see line 66 for example.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Overall Comments
The work is clearly better than the first version. However, there are still some adjustments that should be made. Below are the main points to be considered.

Major Reviews Round 2
1 - The abbreviations “ROI”, “CFAR”, “PV”, and “PRDV” were first defined in lines 12, 67, 175, and 177, respectively. These abbreviations were then defined again in the caption in Figure 2. These abbreviations should be defined once.

2 - I didn’t understand why sometimes in the text you consider “????” and other times you call it “?????, ????? or ????? (calm, interference, land). Why do you differentiate it?

3 - PFA is a commonly used parameter at work however I haven’t seen any definition of PFA.

4 - It is necessary to correct the caption in Figure 9. “Recognition” is misspelled and it is repeated.

Minor Reviews Round 2

1 - In line 146, you should include a “space” between ??, ??, and ??.

2 - In line 160, when citing the work of An [22], it is necessary to write the numbering of reference to avoid future confusion with “an". I think in this case you can just use [22]. Please review that in the whole document. 

3 - In line 198, you should put a comma before the word “respectively”.

4 - I noticed in the text that you called ????. However, I didn’t see any definition. You should define it before use it.

5 - ???? in line 399 should be written as (???? ).

6 - FAR has already been defined in the text and you do not need to define it again in the caption of Table 9.

7 - ?ℎ?? e ?ℎ???? have already been defined in the text so you don’t need to define them in the caption of Table 4.

8 - You shouldn’t define “ROI” in the caption of Table 6. This has already been presented.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Some comments formulated during my review are presented below. These are as follows:

-Unfortunately, the reviewer see very little scientific novelty in this paper. 

-The authors use classical methods for their proposal, I do not see any improvement.

-What are the parameters and their justification for the filters? 

-Sensitivity to parameters is not discussed.

The defects found in the article and presented above represent only some of its shortcomings. The paper has much morer.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

The revised version shows some improvements, however, in my humble opinion, the paper does not convey any relevant methodological advance.

Considering these pieces of information, I suggest the Academic Editor to evaluate if the paper is suitable for publication in Remote Sensing or if it can be better published elsewhere.

Other remarks.

(1) The Authors now provide some numbers, but the numbers are not well motivated or justified. They are still lacking details that would allow reproduction of their work. The Authors should present sufficient detail so as to allow reproduction of their technique and results.

(2) The Authors did not indicate what percentage of the data set was used for training or parameter optimization of the methods. According to good practice, no image that was used in system design and in selection of its parameters should be used in the testing of the systems.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf