A Novel Waveform Decomposition and Spectral Extraction Method for 101-Channel Hyperspectral LiDAR

Round 1

Reviewer 1 Report

This paper proposed a method specialized for multi-channel waveform recorded by HSL, which can improve the correctness of decomposition compared to the single channel waveform decomposition method. However, there exists several points need to be taken into consideration and must be carefully considered and responded with detailed illustration before a further judgment.

 1. This paper lacks comparation between performance of existing multi-channel waveform decomposition methods (which is treated as a baseline of hyperspectral lidar waveform decomposition) and the proposed method. The resisting methods includes the waveform accumulation (waveform selected from the whole channels based on their signal quality (SNB)), at least. Without this comparation, it’s hard to illustrate the improvement worked (proposed) by this paper. It’s strongly suggested to complete the comparation.

 2. The instability of emission of laser energy (every single waveform) has a great impact on spectra detected by HSL. As this paper mentioned the unchanged spectra of the target from the camouflage network, the calibration of the effect of emission of laser energy at the whole band should be considered firstly. Thus, it’s strongly suggested to point out the calibration of this components or illustrate it clearly.

 3. In the part of conclusion, the effect of the proposed method on improving the effeteness should be included, as this paper has discussed the number (101) of channels of this HSL system. Thus, giving a conclusion about it is strongly suggested.

Author Response

Dear reviewer:

  Thank you very much for your careful reading of our manuscript, it must have taken you a lot of time. We have carefully read your comments, and made some modifications correspondingly. We have also carefully revised the grammar and made English revisions by using MDPI’s editing services. We hope that the revision is acceptable, and we look forward to hearing from you soon.

  Please see the attachment.

  Please 

With best wishes,

Yours sincerely,

Yuhao Xia

[email protected]

Author Response File: Author Response.docx

Reviewer 2 Report

The English language of this manuscript should be improved by a native speaker.

Is the high-speed acquisition card with sampling rate of 5 GS/s used in the system a commercial product? What is the storage depth of this card?

As the system covered a large range of wavelength for 500nm, the differences of transmission rates of both transmitting and receiving optical components have to be considered and well calibrated.

How to eliminate the background light transferred into the receiving system?

The FWHM value in Table 1 is missing.

How long does it need to finish a turning from 550-1050nm? Are the wavelengths turned continuously or step-by-step? If it’s a step-by-step turning, how long is the time interval of wavelength switch and how long is the time for measuring at each wavelength?

If it’s continuous turning, the first and last so-called channels have different wavelength coverage (2.5nm) while in other channels it’s 5nm. Therefore only 99 channels are with identical wavelength coverage. Or, if all the channels have the same wavelength coverage range of 5nm, the total number of channels should be 100. Anyway, strictly, it should not be 101 channels.

For a target with a certain length or depth, the echo signal with different wavelengths will have an overlap, which will mix the intensity information from different part of the target as well as decrease the spectral resolution. How do you deal with this issue?

The labels in figure 2 are wrong.

The y-axis in figure 4b is wrong.

It’s stated in line 167 that all channels with a maximum intensity value of less than 4mv are eliminated. Why in figure 4c those signals are still shown? This should be clarified to avoid misleading readers. I suggest to show the fitting results of points eliminated in figure 4b as well.

Author Response

Dear reviewer:

  Thank you very much for your careful reading of our manuscript, it must have taken you a lot of time. We have carefully read your comments, and made some modifications correspondingly. We have also carefully revised the grammar and made English revisions by using MDPI’s editing services. We hope that the revision is acceptable, and we look forward to hearing from you soon.

  Please see the attachment.

  Please 

With best wishes,

Yours sincerely,

Yuhao Xia

[email protected]

Author Response File: Author Response.docx

Reviewer 3 Report

The authors present a good overview of previous approaches to hyperspectral full waveform lidar analysis in the introduction and clearly state their research aims. However, from the text and figures in the Materials and Methods and Results sections I do not believe they have clearly demonstrated that they have addressed those research aims.

My general comments are as follows:

1. The description of the FWHSL system itself is very limited and does not give the details necessary to interpret the results that follow it.  For example, what is the power spectral density of the supercontinuum laser source? What type of detector is used and what are the noise characteristics? What is the expected performance based on the target distance? Is there any spectral filtering on the receiver system that would reduce the effect of passive light sources?

2. Similar to the instrument description, the experiment section is very short and give very little information about why the test is designed in this way? What is the goal of testing a covered target? Is the goal to measure spectral data from under canopy coverage? Tying this experiment into the eventual use of this instrument for scientific investigations is missing entirely. There is no information about what the targets are or what the expected spectra might be.

3. The discussion of the filtering of channels based on the various fitting parameters is the clearest part of the manuscript, but there is very little description of why or how the PCA is done and why it is a better approach than simply comparing the 70-wavelength spectra that were obtained from the decomposition steps.

Please see more specific comments in the attached document.

Comments for author File: Comments.pdf

Author Response

Dear reviewer:

  Thank you very much for your careful reading of our manuscript, it must have taken you a lot of time. We have carefully read your comments, and made some modifications correspondingly. We have also carefully revised the grammar and made English revisions by using MDPI’s editing services. We hope that the revision is acceptable, and we look forward to hearing from you soon.

  Please see the attachment.

  Please 

With best wishes,

Yours sincerely,

Yuhao Xia

[email protected]

Author Response File: Author Response.docx

Reviewer 4 Report

The Paper entitled" A Novel Waveform Decomposition and Spectral Extraction 2 Method for 101-channel Hyperspectral LiDAR" presents a very interesting approach to optimize the spatial and spectral identification o distributed targets in such a way as to reduce  cases of misidentification.   Th approach is clearly complex and may be hard to implement but it is a useful approach to learn and may fimd an increasing number of applications 

 

Some issues 

1) The flow chart diagrams are hard to read because the authors place them side by side. I think it is necessary to put each half of the flow chart on a separate page and make the direction arrows "broken" so that it is still possible to see the logical flow of the algorithm. 

 

2) It seems the authors have not removed the template instructions "see lines 360-366

 

3) It is not clear how this complex multi-level algorithm performs time wise. I think the authors should discuss CPU time comparisons 

 

Finally, the English is often quite awkward (although it is understandable with effort) and the paper needs careful proof reading from an English Speaker or Professional Editing service. 

 

Author Response

Dear reviewer:

  Thank you very much for your careful reading of our manuscript, it must have taken you a lot of time. We have carefully read your comments, and made some modifications correspondingly. We have also carefully revised the grammar and made English revisions by using MDPI’s editing services. We hope that the revision is acceptable, and we look forward to hearing from you soon.

  Please see the attachment.

  Please 

With best wishes,

Yours sincerely,

Yuhao Xia

[email protected]

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Most of the questions has been figured out.

Reviewer 3 Report

I thank the authors for their careful responses to the review comments and modifications to the manuscript. I believe the revised manuscript is much improved and suitable for publication.