Vector Magnetic Anomaly Detection via an Attention Mechanism Deep-Learning Model
Round 1
Reviewer 1 Report
- The paper is very good with a clear novel idea. However, I have some insights to increase the impact of the paper.
- Figure 1 is clear and good. However, for anyone unfamiliar with UNET this figure will be very difficult to interpret. I do suggest the authors redesign it as it is the most important figure in the paper.
- The number of experiments is few. I do suggest the authors do more experiments to enrich this paper.
- I do want the authors to discuss more the RNN and to explain it and how it was incorporated in their model.
- The authors must explain more about the UNET and give more details.
- The literature review is missing! The authors must discuss recent literature especially in the usage of UNET even in other domains.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
It was complicated for me to understand the major points of this paper.
The number of things that have to be clarified clearly and concisely:
- What was the purpose of this work? Why is it innovative regarding other works done before?
- What exactly was done by the authors?
- Describe In physical terms what is the magnetic anomaly detection
- Give a simple explanation of the deep learning process regarding magnetic anomaly detection.
- What can I learn from the graphs in this paper?
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors did an amazing work. The quality of the paper has increased. However, I want the authors to present the results in a table. I am still confused by the results section. The paper must be rich in experiments.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 2 Report
The authors answered all my questions and remarks. Now I'm sure that this manuscript is ready for publication.
Author Response
Thank you very much for your comments on our manuscript.