Deep ResU-Net Convolutional Neural Networks Segmentation for Smallholder Paddy Rice Mapping Using Sentinel 1 SAR and Sentinel 2 Optical Imagery

Round 1

Reviewer 1 Report

The article contributes to the Remote Sensing journal, however is important to add some comments that I alreday did to the articlce body.

In all manuscript please write (i.e., vegetative, reproductive and ripening) (homogenize).

Please write equations 4 to 7 instead of 1 to 4.

In Figure 5, In the legend please write Jiguan instead of Jiaguan District.

In the Discussion section please add one or two discussions with other researchers (compare findings).

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

I found this paper interesting work and well developed research. The idea of research is quite interesting and I assume the manuscript will benefit some minor revise as following.

First, The paper has to be reformatted according to the journal’s format.

I would suggest to combine the research questions with the text rather than numbing which makes paper similar to proposal rather than scientific paper, , perhaps you could use research questions are a), b) and etc..

Right now it is a bit difficult to figure out how methods are applied and combined as approach since there are variety of methods and techniques used in this study, I would suggest create a new section after study site and discuses overall methodology of the research perhaps you can consider it ins several steps and explain each step in brief, you can also address the step in figure 2 which I believe will help readers to figure out the methodology and its combination efficiently.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

1. The detailed introduction of the workflow in the article can make readers understand the main work of the article well. But for this work, the methods used in the work are all existing models. The method of combining optical and SAR data used has also been studied a lot before, and the author has not further mined the data and used or proposed new available parameters on the basis of these studies. Neither the methodology nor the data show the author's own contribution. The highlights of the article should be shown in your manuscript.

2. Neither the characteristic of rice nor the method principle are discussed about the reason of the high accuracy of the optimal U-Net predictions. And some figures are only shown in the manuscript, and not further analyzed in the result or discussion (such Figure 4).

3. This paper gives a detailed introduction to the indexes which used to evaluate the experimental results. These indexes are calculated based on the confusion matrix, but the specific values of the confusion matrix are not given in the paper. The author can attach the table of confusion matrix.

4. There are two tables of table5 in the article, and table4 is missing.

5. The date of the images, which are used in the manuscript, should be list in a table.

6. “An independent set of ground validation data was generated using a stratified random approach for accuracy assessment.” is in the Section2.2.4. The stratified random approach should be in more detail.

7. Are the training data of RF classification and the optimal U-Net predictions different? Please specify explicitly.

8. “Table S3 presents the accuracy assessment results associated with the upscaled paddy rice mapping on the district scale for 2016, 2018, and 2020, respectively.” is in the Section 3.3. Are  the ground validation data of 2016 and 2018 the same as 2020. Are there some samples land cover of which are changed? Please specify explicitly. 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

On behalf of my co-author, I want to thank you for all the valuable edits provided in our manuscript. 

All the comments raised were updated in the revised manuscript.

Please see the attachment.

Thank you

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The manuscript have been revised well. I hold that the paper can be published in present form.