Object Detection via Gradient-Based Mask R-CNN Using Machine Learning Algorithms

Round 1

Reviewer 1 Report

The authors seem to have made sincere attempts to address the review comments from the previous round of reviews. Thus, in my opinion, subject to the following minor corrections, the manuscript may be accepted for publication:

1. The authors should clarify somewhere in the Introduction section that their proposed method does not relate to YoLo, as they mentioned in their response "Because the main core of this research paper is based upon RCNN and not related with YOLO." Otherwise readers moght wonder if the authors are unaware of any possible perceived similarity of their proposed method with a popular one such as YoLo.
2. When the authors describe the image quality paper ( reference 21 ) in the last paragraph of Related Works, they should add a few sentences on its potential use to study the relation between image quality and object detection performance as future work.
3. The manuscript still needs some more English editing to correct errors in phrases like the one on Line 67: "demonstrate its tremendous potential to **meet** a variety of obstacles"

Author Response

 

Reviewer-1 Comments and Responses

 

Comment -1:

 

The authors should clarify somewhere in the Introduction section that their proposed method does not relate to YoLo, as they mentioned in their response "Because the main core of this research paper is based upon RCNN and not related with YOLO." Otherwise readers moght wonder if the authors are unaware of any possible perceived similarity of their proposed method with a popular one such as YoLo.

 

Response:

Thank you so much for your valuable time to correct our faults.

Now we corrected and updated introduction accordingly and highlighted in manuscript as follows,

 

 

Comment-2

 

When the authors describe the image quality paper (reference 21) in the last paragraph of Related Works, they should add a few sentences on its potential use to study the relation between image quality and object detection performance as future work.

 

Response-2:

Thank you for your valuable comments on our research work. The following changes that we made in our manuscript. Kindly check it.

 

 

Comments-3

The manuscript still needs some more English editing to correct errors in phrases like the one on Line 67: "demonstrate its tremendous potential to **meet** a variety of obstacles"

 

Response-3:

Kindly find the following changes that we made in our manuscript as follows. Moreover, this manuscript was given to one of our native English speakers to correct errors and mistakes.

 

 

 

Thank you for your valuable comments. kindly apologize if any mistakes or errors.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper deals with some quite common tools so contributions are quite difficult to be found. But authors highlight them and that is a positive point. However, there are some issues that must be improved before publication:

• The title is a little bit confusing and it is not clear the application of what is presented. I suggest revising it to provide a clearer idea of how the method could be used
• Section 2 must be revised and reorganised. A state os the art is not a commented list of references but they must be organised in groups.
• Sections 3 and 4 could be merged.
• In sections 5 and 6, there are long explanations before a subsection title. A subtitle at the beginning would improve understanding. 

Author Response

 

Reviewer-2 Comments and Responses

 

Comment-1

The title is a little bit confusing and it is not clear the application of what is presented. I suggest revising it to provide a clearer idea of how the method could be used.

 

Response -1:

 

We changed the title to avoid confusion as follows,

 

Comment-2

Section 2 must be revised and reorganised. A state os the art is not a commented list of references, but they must be organised in groups.

 

Response-2:

We reorganized related works (Section 2) and set up Table 1 according to your comments.

 

Comment -3:

Sections 3 and 4 could be merged.

Response-3:

We merged Section-3 and Section 4. Thank you for your comments and suggestions.

 

Comment-4:

In sections 5 and 6, there are long explanations before a subsection title. A subtitle at the beginning would improve understanding.

 

Response -4: Thank you very much for your suggestions. The following changes that we made in our manuscript. Kindly find it.

(i)
(ii)	In Section 4 (now updated), we updated as follows
(iii)	In Section 5 (now updated), we updated as above.

 

Thank you so much for spending your valuable time.

Kindly apologize if any mistakes or errors.

 

Thanks again.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Some of my previous queries have been taken into account but some issues have not been addressed:

• The state of the art remains as  list of references. The information must be grouped and not simply include a paragraph for each paper.
• It is not necessary the last change of including a mini-index of each section but including a subtitle at the beginning of the largest sections

Author Response

Reviewer Comments and Response

Comment-1

The state of the art remains as  list of references. The information must be grouped and not simply include a paragraph for each paper.

 

Response-1:

Thank you for your comment. The following changes that we made in our manuscript. Please apologize if anything missed or errors.

 

Comment-2

It is not necessary the last change of including a mini-index of each section but including a subtitle at the beginning of the largest sections.

Response -2

Dear Sir,

We removed the mini-index part from manuscript as it not necessary to include. Hence, we mentioned the following subtitle at the beginning of the sections 4 and 5.

Again, if any mistakes or errors, please apologize.

 

 

Thank you for your valuable time to correct our faults.

 

Author Response File: Author Response.docx