Towards Edge-Precise Cloud and Shadow Detection on the GaoFen-1 Dataset: A Visual, Comprehensive Investigation
Round 1
Reviewer 1 Report
The manuscript is devoted to interesting and importatnt scientific-technical problem - the spectral remote sensing images processing and identification of the signal against the obstacles like clouds and shadows.
Abstract contains abbreviations, but some of them are not described, like CRF, OLI.
UNet abbrevaiton needs explanation, especialy for those how are not familiar with this technique.
The introduction section is not well organized. There is no explanation of the study goals.
Fig. 1 is not well described in the text of introduction section.
Line 49 - GaoFen (GF) abbrevation explanation is not needed, as abstact already provides it.
CRF is explained in line 100, bet first mentioned in 18, this should be corrected.
The literature review in section 2 could be improved by extending brief description of the available methods, their disadvantages and the proposed method advantage. Despite authors put literature references some brief description of the mentioned deep learning techniques is desirable. This can allow readers who are not familiar with described techniques to understand better the further content of the manuscript.
Section 3.1 provides mathematical model of the analysis, however it's not easy to understand the sense of the equations for those who are not familiar with this technique. Some more detailed explanation fo used formulas is advisable.
I suggest to give brief essence of segmentation method which is intesivey mentioned in the manuscript.
It would be possible to add to the captions of Figures 2 and 3 the time, date and place of the images, as well as the channel names.
In general, the article is of scientific interest to readers. The proposed optimization of methods for filtering satellite data of various types seems reasonable and sums up the results of the study. The article can be accepted for publication after revision.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Please find attached comments and suggestions for authors.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors generally did the necessary work to improve the article according to our recommendations. In this form, the article can be accepted for publication.
