Multiscale Superpixel Guided Discriminative Forest for Hyperspectral Anomaly Detection

Round 1

Reviewer 1 Report

The paper is of interest. However, it needs to be revised before publication. 

The presentation of figures is not good. The introduction is very long and can be separated into two sections. 

It would be great if the authors evaluate the sensitivity of the feature selection. This can be an additional table or figure. 

On the number of trees, what is the effect of adding a cost function to improve the results and the accuracy? 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear authors,

 I want to let you know that I appreciate your substantial efforts in this impeccably and professionally presented research. This paper promises excellent visibility as it has great potential to capture the attention of readers and specialists working in the field. The theme developed is fascinating in terms of scientific content and practical applicability. To further increase the overall quality of the manuscript, I would like to address the following comments and suggestions:

1. For easy reading and understanding of the content, inserting a table with all abbreviations and acronyms in the text is helpful.

2. For implementers, it is also welcome to mention the software package and its version used for all simulations.

3. Please briefly describe the building blocks of the proposed method shown in Figure 2.

4. In the last section, Conclusion, please mention some research directions for your future work.

Thanks, 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Good for the publication in the present form

Author Response

Please see the attachment.

Author Response File: Author Response.docx