A Deep Learning Localization Method for Acoustic Source via Improved Input Features and Network Structure

Round 1

Reviewer 1 Report

The manuscript presents a deep learning localization method via improved input features and network structure, which can effectively estimate the depth and the closest point of approach (CPA) range of acoustic source. The results in this paper in promising. I think that the manuscript is interesting and scientifically sound. However, before the decision of acceptance is running, some corrections should be addressed. Please find the attached file for details.

Comments for author File: Comments.pdf

The English Language should be improved.

Author Response

  Thank you for your professional review work on our article. We have revised the manuscript based on your suggestions. The modification instruction has been uploaded as a PDF attachment. In the attachment, we have addressed each of your questions individually and indicated the modifications made in the manuscript. Thank you once again for your positive comments and valuable suggestions to improve the quality of our manuscript.

Author Response File: Author Response.pdf

Reviewer 2 Report

I would like to congratulate the authors for the interesting and innovative manuscript. However,  I leave you here some comments and suggestions:

Line 12: prove the effectiveness.

Keywords: I would suggest avoiding repeating terms (e.g., deep learning, acoustic localization) already included in the title and/or abstract in order to increase article findability in search engines.

Introduction section would benefit from an English proof reading from a native speaker colleague or a professional traductor to increase readability and linkage among certain sections/phrases.

Line 28, 42, 46, […]: I am not sure if Remote Sensing admit this kind of citation (Chen et al., without the [ref. number] until the end of the phrase). Usually is presented just following the “et al.,”. If it is not admitted, please correct along the manuscript.

Line 31-34: The statements “First, the localization performance is heavily dependent on the environment parameters and has limited resistance to environment mismatch, which makes the MFP methods difficult to apply in the complex ocean environment. Second, the improved algorithms usually require a large amount of running time.” Has no reference. Please add one/more than one.

Line 45: the phrase “The Noise09 experiment further verified the conclusion” it is not clear. I would suggest rephrasing and integrate into the previous phrase.

Line 47: Please add the meaning of the acronym the first time is mentioned in the text. The same issue repeats again in Line 49 for Convolutional Neural Network, which could be already presented in Line 40.

Line 48: “The SWellEx-96 experiment verified the ability of the method”. I would suggest rephrasing and integrate into the previous phrase to increase readability.

Line 74: Aiming at solving?

Line 76-96: I would avoid some repetitions in this section to increase readability.

Line 107: I am not familiar with this kind of manuscript structure (sections 2 – 4), if the Journal allow it, please ignore this comment. If the Journal do not accept it I would suggest to cluster together from 2-4 in a Methods section, 2-3 in one subsection and 4 in another one.

Line 149: “Fast Fourier Transform”

Line 239: “The convolution module consists of”

Section 4.2, 4.3 are phrased in present in contrast with the rest of the manuscript that is generally written on past tense. Also, some subsections of section 5 there are some changes between past tenses and present.

Line 301: Please include the reference for NVIDIA (NVIDIA Corporation, Santa Clara,
CA, USA).

Line 407: I would rephrase “further prove it” to “back up” this statement.

I think some sections could benefit from the proof reading of a native English speaker to homogenize and increase readability. Specially Introduction and some sections highlighted in the Comments section regarding verbal tenses.

Author Response

  Thank you for your professional review work on our article. We have revised the manuscript based on your suggestions. The modification instruction has been uploaded as a PDF attachment. In the attachment, we have addressed each of your questions individually and indicated the modifications made in the manuscript. Thank you once again for your positive comments and valuable suggestions to improve the quality of our manuscript.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

There's no problem any more.