Smart Decision-Support System for Pig Farming

Round 1

Reviewer 1 Report (New Reviewer)

This is a well written manuscript.

However, more experimental results will help for readers to understand the proposed method.

Also, in Table 1, authors need to check mAP 72.4, 0.5IoU 97.1, and 0.75IoU 87.7 with Resnet50. Why was the mAP value much worse than the IoU values?

Author Response

Please see the attachment, thank you.

Author Response File: Author Response.pdf

Reviewer 2 Report (New Reviewer)

In this paper, the authors have proposed an intelligent decision support system for swine farming. Overall, the manuscript is well written but requires some important improvements which are mentioned below:

1) In the introduction section more needs to be explained about the following sections of the article.

2) Please explain more about your unmanned vehicle prototype, including its characteristics and some material that other authors can use to reproduce your experiment.  

3) Include a comparison table after the related work section that should highlight the strengths and weaknesses of the proposed method.

4) A results section should be included, including a comparison of the metrics of related work with your results. 

5) You have collected your own data set for the experiments; provide the link to the data set to reproduce and verify the results. 

Author Response

Please see the attachment, thank you.

Author Response File: Author Response.pdf

Reviewer 3 Report (New Reviewer)

This paper proposes a smart decision-support system for pig farming. The proposed method has certain application value. Recommendations for improving the manuscript:

1.      Where does Figure 1 come from, please quote.

2.      There is a lack of experimental comparison of the latest semantic image segmentation methods.

3.      The operation efficiency of the algorithm has not been analyzed.

Author Response

Please see the attachment, thank you.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report (New Reviewer)

The manuscript was well revised.

Reviewer 2 Report (New Reviewer)

Accept. 

Reviewer 3 Report (New Reviewer)

The authors have well addressed my previous concerns, thus I accept this revised version.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

I've revised the paper Drones-1915627. While the paper covers a potentially interesting topic, I have too many concerns regarding the insufficient details given about the content of the work presented.

The manuscript it is not a scientific paper as it only presents general ideas and a description of some implemented system without providing details to replicate nor giving evidences of a fair operation.

 

There is a clear lack of suitable references with published papers dealing with pig segmentation and weight estimation. There are even several reviews dealing with this topics not cited in the text, se for instance:

 

Jiangong Li, Angela R. Green-Miller, Xiaodan Hu, Ana Lucic, M.R. Mahesh Mohan, Ryan N. Dilger, Isabella C.F.S. Condotta, Brian Aldridge, John M. Hart, Narendra Ahuja, 2022. Barriers to computer vision applications in pig production facilities, Computers and Electronics in Agriculture, 200, 107227

 

Similar references can be found in MDPI journals like Sensors or Animals

 

To this respect, the improvement and contribution of the proposed paper is not provided with regard existing publications.

 

There are methodological issues regarding the development of the deep learning techniques involved in the farming support system like:

-         The scarce use of images available for training and validation

-         The lack of specification of video camera characteristics

-         Details of pretraining models

-         The lack of performance measures to assess the correct operation of the system in pig segmentation or pig weight estimation

-         Abounding on colloquial, general and unspecific terms instead of providing proofs based on measures: “good performance” or “satisfying results” or “can”

-         Experimental results are clearly insufficient

For these reasons, I strongly recommend the rejection of the paper and I suggest the authors to re-write the paper in deep, with more detail providing scientific details and evidences of the results after presenting a good description of the material and methods used in this project.

Reviewer 2 Report

Dear Authors and Editors:

I would like to send the review report.

Please find the attached file in this email. 

Thank you.

Sincerely yours.

The reviewer.

 

Comments for author File: Comments.pdf

Reviewer 3 Report

The article presents the "Smart Pig Farming Support Systems With Digital Twin". The article lacks the knowledge of digital twins and lacks the experimental results and evidence to the several hythoses. However, the following comments will improve the quality of article.

1. An abstract is unclear. Rewrite the abstract by considering a brief overview of the problem, proposed framework, and significant achievements.

 

2. The authors should provide proper references for paragraphs 1, 2, and 3 of the Introduction.

3. In line 46, the authors mentioned poorer nations the word poorer is awkward. Change it to the formal word.

4. The title of the article is depicting that the article uses digital twin but there is no explanation and experimental evidences about digital twin are presented in the article. It seems like the authors misinterpreted the term “digital twin”.

5. For image stitching the authors used only SIFT features. However, there are other feature descriptors such as AKAZE and YAPE. The authors must compare the performance of SIFT with others.

6. Rewrite the conclusion as it is not coherent with main body.

7. How the authors addressed the issue of stitching failure?

8. The role of drone must be elaborated in a separate section. It will be better if the authors add the sample images taken from drones.

9. The abbreviations and acronyms must be explained when used first time in the article.

10. In subsection 3.3 the authors mentioned the weight estimation of the pig alleviates the risks of the illness. This sentence is not clear and must be elaborated with proper justification and evidence. Authors must discuss how the weight of a pig is related to the illness.

11. There is no information about digital twin (DT) and how the DT is simulated, how it is modeled?

12. What is the significance of this study?

13. What are the limitations and future directions of this system?

14. The proposed methodology and experimental results sections are very weak and miss some critical details these sections must be improved.

Thanks, and Regards

The Reviewer