Enhancing Forest Security through Advanced Surveillance Applications
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors1. Need to add the source of this figures.
2. Introduction is very long and content many different information. Is better to shorten and you add need subsection for some these information.
3. (2.3. Radar and Camera Fusion) what you need to say?
4. Figure 5. (Algorithm Flow Diagram) better to select and better title for this figure. (System block diagram)
5. In section 3.1 for image tracking, can you mention the important features used in this detection.
6. You didn’t mention figure 6.
7. Line 347 (Following paragraphs will concentrate on description how the result were achieved 347 without describing all blocks of the VMD.) is not clear.
8. Need to number all equation and be sure to mention all symbols.
9. Line 589 the equation in what?
10. The paper is too long and content many details is better to divided into papers and focus more important novelty.
Comments on the Quality of English LanguageModerate editing of English language required
Author Response
Hello,
Thank you for your comments and suggestions brought by you.
Attached please find a replay for issues raised by you.
B.R
Danny
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsTitle: Enhancing Forest Security through Advanced Surveillance Applications
Authors: Danny Buchman, Tomas Krilavičius, Rytis Maskeliūnas
Summary:
The article claims to present a multi-sensor system for monitoring forests.
It uses camera tracking methods to track multiple targets and convolutional
networks for object classification in images. It also appears to use
a thermal camera and an MMW radar fusion model.
General Comments:
I found the article very confusing. It is too long and lacks focus.
I was unable to determine which sensor contributes to the overall
performance of the system. While multiple sensors are necessary,
it is also necessary to evaluate the contribution of each sensor.
Table 5 at the end of the paper lists a number of events that the
system is supposed to detect. I could not tell from the text of the
article how many of these events were used for training the system
and how many were used for testing/validation. What was the dataset
on which the system was evaluated?
Was the system actually deployed in a target environment? Figures 16, 17
suggest to me that at least some parts of this system were deployed only
in a laboratory. If that is the case, then the article proposes a research
prototype, which should be clearly stated in the abstract.
The authors should focus on a smaller, more manageable subset of issues,
e.g., hardware design and deployment for data capture. Sensor fusion
methods could be addressed in a future publication after a comprehensive
dataset has been curated from a deployed system.
Specific Comments:
1. Could you briefly define the term "afforestation"? Does the term mean
re-planting deforested areas?
2. The text of the article should be re-grouped into longer paragraphs.
For example, the paragraph in lines 26-29 contain only a single sentence.
The previous and subsequent paragraphs contain 2 sentences.
3. Lines 37-38: "Monitor wild animals that are on the premises and,
if necessary, close access to humans in case of need."
This sentence must be re-phrased. Did you mean to say "We would like to monitor wild animals..."
4. Figure 2 and 3 should state the GPS coordinates and location of these reserves?
5. Lines 82 -- 83: "Among these, cameras, radar and LiDAR have frequently
used options, each with its own advantages and disadvantages."
Change "have" to "are."
6. Table 1: The caption to the table should explain the color coding scheme in the table.
7. Line 263: "Main achievements are as following:"
Re-phrase this sentence as "Our main contributions are as follows:"
8. Line 265: "requered" --> "required."
9. The equations in Section 3.2 should be numbered and referenced in the text.
10. Lines 717-720: "Therefore, we focused on improving the accuracy of our VMD detector, which would
help us detect moving objects more reliably. At the same time, we worked to fine-tune the
detector’s sensitivity, so that it could still detect targets that were not classified without
triggering false alarms.
I did not find much evidence in the paper to support these claims. What were the testing dataset?
What were the training dataset? Did you use Precision, Recall, IOU, F1? How do you know that
your system works.
Figures 16, 17 suggest that this part of the system was deployed only in a laboratory. Is that
correct?
11. Lines 747-750. "Our client’s requirements were focused on achieving low false alarm rates and high
detection probability. While the client did not provide an explicit definition of the prob-
ability of detection, it was evident that the system user was most interested in the early
detection of objects of interest entering the observed area."
Who was the client? A government agency? A private company?
The article requires extensive editing.
Author Response
Hello,
Thank you for your comments and suggestions brought by you.
Attached please find a replay for issues raised by you.
B.R
Danny
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors presented a combination of advanced technologies, such as radars, thermal imaging, remote sensing, artificial intelligence, and biomass monitoring systems, in the field of forestry and natural resource security. The idea discussed in this paper is very interesting, however, I have the following comments that the authors should address:
- The paper is lengthy; the authors should be more concise and provide the most important concepts to keep readers focused on the paper's main contributions.
- Improve the quality of figures in the paper.
- Give numbers to all equations and list them in the paper.
- Line 536, I didn’t understand the meaning of Pk|k. What do you mean by the posterior probability of the parameter k given the evidence k? The authors are requested to check the equations where they have used k|k.
- Algorithm 2 should be written in a better proper way. Put steps one per line.
- Line 629, correct the sentence; there is “14”, maybe is should be as depicted in Figure 14.
- There are many English errors and typos that should be addressed (e.g., “Figure 16 and 17” should be “Figures 16 and 17”, “Two test setup were build, calibrated and tested” should be “"Two test setups were built, calibrated, and tested.", ...)
- The authors should integrate this recent important paper that discusses the fusion concept in their work: 10.1016/j.inffus.2023.102006
- The experiments would benefit from comparing the impact of fusion in the accuracy. I mean here, including an ablation study to show the performance of the system using fusion and without fusion.
Author Response
Hello,
Thank you for your comments and suggestions brought by you.
Attached please find a replay for issues raised by you.
B.R
Danny
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsNo more comments
Author Response
Thank you for reviewing and accepting the performed job.
Reviewer 2 Report
Comments and Suggestions for AuthorsDear Authors:
Thank you for taking time to address my comments. The article is greatly improved. The quality of the English language is much better. It reads
well. The quality of the figures and captions is much better.
I still think it is too long and reads like a book chapter, not a journal article. It is 37 pages long: 33 pages of content and 4 pages of references.
As you state in Line 531, the dataset is proprietary. An immediate implication is that your results are not replicable. In other words, other researchers, such as myself, cannot reproduce your results, even in principle. Thus, I cannot, in good conscience, as an open science
researcher recommend your work for acceptance. It is for this reason that I indicated that your research design must be improved.
If you cannot make the entire dataset public, perhaps you can make parts of it available through the supplementary materials. My understanding of the MDPI publishing guidelines is that the datasets used in open access publications must be publicly available.
Here is a question for you to ponder. Why do you want to publish your results in an open access journal? Why not go for an Elsevier or Taylor & Francis journal? Elsevier and Taylor & Francis are known for putting
access paywalls to their publications. Perhaps, these two publishing organizations are more appropriate dissemination venues for your research.
Specific Comments:
1) Line 22: Correct the figure reference in Figure ??.
2) The formulas b/w the lines 475 and 476 need to be
adjusted so that the comment does not go over the
right margin.
Author Response
Thanks for your review. Please refer to the attached replay letter
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors addressed my comments. The paper can be accepted.
Author Response
Thank you for reviewing and accepting the performed job.
