An Optimized Semi-Supervised Generative Adversarial Network Rice Extraction Method Based on Time-Series Sentinel Images

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Manuscript entitled “An optimized SSGAN rice extraction method based on time-series sentinel images” focuses on proposing an optimized SSGAN rice extraction method based on time-series Sentinel-1 SAR images and Sentinel-2 multispectral images, to achieve high-precision rice extraction while alleviating the demand for labeled samples to a greater extent in addition to an innovative focal-adversarial loss function designed to encourage the discriminator to consistently and effectively approximate the gradient updates of the generator network. Overall, the manuscript is well prepared. There is a need for improvement in scientific language. Some specific comments are mentioned below to be addressed.

-          Please check and correct the citations thought MS.

-          What is your opinion on weather/climatic conditions on satellite visit? How does it impact the quality of the images and data?

-          Sentinel-1 has 12 days for revisiting the same spot. In the table 1; sentil 1 , ….. sentil 18 means the number of visits?

-          I think it is important to provide the weather data including cloud factor and explain the variations on each visit. Evaluate and explain of the imaging was affected.

-          How many total samples were used for training and validation?

-          Discussion is not sufficient. Please expand on how this study is advantageous using ML.

-          Please shorten the conclusion. Mention the key findings in conclusion only.

Comments on the Quality of English Language

Moderate editing of English language required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In my opinion, the section Study Area and Data should be included in general section Materials and Methods.

Line 312 – why did you use NDWI instead of NDMI? As far as I know, NDMI is better indicator of plant health, while NDWI is used primarily to determine the water bodies boundaries.

Methodology for the Mean Intersection over Union (MIoU) and Overall Accuracy (OA) indicators should be added in the Materials and Methods section of the article.

Overall accuracy of rice extraction rarely differs more than 5% by the data amounts and methods. Is it fair in this case to tell what method is superior to the others? Additional statistical prove for the significance of these differences would be beneficial.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for addressing comments and revising the manuscript. 

Comments on the Quality of English Language

Minor editing of English language required.